{ "complete": true, "report": { "bucket": "vn_S_RANT_reg", "label": "Strong Ranting/Angry Style", "definition": { "id": "vn_S_RANT_reg", "category": "talking_style", "head": "vn_S_RANT_reg", "pole": "high", "label": "Strong Ranting/Angry Style", "floor": 4.0, "canonical_dimension_name": "Ranting/Angry Style" }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/style_s_rant/full-v2-sealed/sealed-selected", "mission": "style_s_rant", "format_status": "CANONICAL", "heldout": true, "qualified": true, "source_cfg": 3, "source_model": { "label": "Humaneness 1.5 + real-speech rank 4", "repo": "laion/Humaneness-Voice-Large-1.5-RealSpeech-LoRAs", "revision": "fed33e554f094fcdc89fc0124aed2f9d42018714", "rank": 4 }, "metrics": { "ce": 6.055615367889405, "genuineness": 3.7212730860710144, "wer": 0.00441296134844522, "identity": 0.5887842419743538, "voiceclap": 0.09205937270075083, "fitness": 0.731840998555987, "steering_v2": 0.8576362198505612, "raw_target": 3.037348734140396, "target_cdf": 0.8576362198505612, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.0064105129241945, "genuineness": 3.685123085975647, "wer": 0.00578785236849753, "identity": 0.5887842419743538, "fitness": 0.7314257163153465, "steering_v2": 0.8644409801381889 }, "none": { "n": 50, "ce": 6.104820222854614, "genuineness": 3.7574230861663818, "wer": 0.003038070328392909, "identity": null, "fitness": 0.7322562807966276, "steering_v2": 0.8508314595629334 } } } } ], "selected_wording": { "id": "manual-9663bdd37f37", "general": "A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.", "inline": [ "A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.", "A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.", "A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable." ], "cfg": 3, "recipe": "P06", "strategy": "initial-physical-directing", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-3", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.6154296537598365, "target_raw": 3.4953970432281496, "ce": 5.701928234100341, "genuineness": 3.6031266331672667, "identity": 0.499321473389864, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.045799288898706436, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.1816406749624486, "target_raw": 2.941908824443817, "ce": 5.854228973388672, "genuineness": 3.708407199382782, "identity": 0.5164889581501484, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.028631804138422014, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.6154296537598365, "target_raw": 3.4953970432281496, "ce": 5.701928234100341, "genuineness": 3.6031266331672667, "identity": 0.499321473389864, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.045799288898706436, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.0302734395984958, "target_raw": 2.7487734854221344, "ce": 5.839733576774597, "genuineness": 3.749358355998993, "identity": 0.5262156374752521, "wer": 0.02042483660130719, "valid_fraction": 0.9, "identity_delta": -0.018905124813318252, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 0.7926391603900498, "target_raw": 2.4455666720867155, "ce": 5.92094738483429, "genuineness": 3.7677157402038572, "identity": 0.5281236737966537, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": -0.016997088491916657, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 0.27587890231760126, "target_raw": 1.7862121731042861, "ce": 6.002359795570373, "genuineness": 3.9291669368743896, "identity": 0.5175030618906021, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.027617700397968292, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": -0.11152344718957642, "target_raw": 1.291910457611084, "ce": 6.042271709442138, "genuineness": 4.0014983177185055, "identity": 0.5431384786963462, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.001982283592224121, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": -0.3784057628307247, "target_raw": 0.9513849377632141, "ce": 6.140165138244629, "genuineness": 4.137060737609863, "identity": 0.5467434346675872, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.0016226723790168762, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.6388427769022991, "target_raw": 0.6190832287073136, "ce": 6.145725059509277, "genuineness": 4.20421439409256, "identity": 0.5635744646191597, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.018453702330589294, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.7464843673724237, "target_raw": 3.662614846229553, "ce": 5.599269247055053, "genuineness": 3.5923004746437073, "identity": 0.46211934722959996, "wer": 0.008350533195734434, "valid_fraction": 1, "identity_delta": -0.08300141505897045, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.1816406749624486, "target_raw": 2.941908824443817, "ce": 5.854228973388672, "genuineness": 3.708407199382782, "identity": 0.5164889581501484, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.028631804138422014, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.1389648531567358, "target_raw": 2.887457084655762, "ce": 5.858001708984375, "genuineness": 3.6737320899963377, "identity": 0.5411741778254509, "wer": 0.015789473684210527, "valid_fraction": 0.95, "identity_delta": -0.003946584463119507, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 0.9024902481475194, "target_raw": 2.585729944705963, "ce": 5.972574758529663, "genuineness": 3.771481382846832, "identity": 0.577399118244648, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": 0.03227835595607757, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 0.7908447130249209, "target_raw": 2.4432770669460298, "ce": 5.958675122261047, "genuineness": 3.7979976892471314, "identity": 0.5312055706977844, "wer": 0.008823529411764706, "valid_fraction": 0.95, "identity_delta": -0.01391519159078598, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.7712890710582134, "target_raw": 2.418325263261795, "ce": 5.919954586029053, "genuineness": 3.8129032850265503, "identity": 0.5190566726028919, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.02606408968567848, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.4644287195664236, "target_raw": 2.0267902135849, "ce": 5.998189854621887, "genuineness": 3.9078283667564393, "identity": 0.583429642021656, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.03830887973308563, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.3625976621714078, "target_raw": 1.896860009431839, "ce": 5.987864398956299, "genuineness": 3.9092405080795287, "identity": 0.5435826316475868, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0015381306409835815, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.32502443160090705, "target_raw": 1.8489188641309737, "ce": 6.0436616897583, "genuineness": 3.923361587524414, "identity": 0.5876842752099037, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04256351292133331, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.1349121107025267, "target_raw": 1.2620679527521133, "ce": 6.09052345752716, "genuineness": 4.053746294975281, "identity": 0.5866998538374901, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04157909154891968, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": -0.1555664132662752, "target_raw": 1.2357143253087997, "ce": 6.091714859008789, "genuineness": 4.082929861545563, "identity": 0.5441830903291702, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.000937671959400177, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": -0.17802734629556669, "target_raw": 1.2070555478334426, "ce": 6.128251147270203, "genuineness": 4.104739105701446, "identity": 0.5199138015508652, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.02520696073770523, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": -0.4193847679105357, "target_raw": 0.8990982323884964, "ce": 6.158234715461731, "genuineness": 4.156830275058747, "identity": 0.5667765468358994, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.02165578454732895, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": -0.4858154259352982, "target_raw": 0.8143367737531662, "ce": 6.153270578384399, "genuineness": 4.14992663860321, "identity": 0.5683604046702385, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.02323964238166809, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.49259032988827517, "target_raw": 0.8056924104690552, "ce": 6.156844806671143, "genuineness": 4.183817219734192, "identity": 0.5327785603702069, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.012342201918363572, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.5970153833131325, "target_raw": 0.672452425956726, "ce": 6.134208154678345, "genuineness": 4.172520339488983, "identity": 0.5987466007471085, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.05362583845853806, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.6212768599291941, "target_raw": 0.6414962649345398, "ce": 6.129243969917297, "genuineness": 4.207979941368103, "identity": 0.5769537329673767, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.031832970678806305, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": -0.7024169883805549, "target_raw": 0.5379664212465286, "ce": 6.164787411689758, "genuineness": 4.248617398738861, "identity": 0.5673681378364563, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.022247375547885896, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.796875016058055, "target_raw": 3.7269102334976196, "ce": 5.542578458786011, "genuineness": 3.545857787132263, "identity": 0.4199021473526955, "wer": 0.008388157894736842, "valid_fraction": 1, "identity_delta": -0.12521861493587494, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.7619140678466025, "target_raw": 3.68230220079422, "ce": 5.617438077926636, "genuineness": 3.5505648255348206, "identity": 0.42967868894338607, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.11544207334518433, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.7601562632113998, "target_raw": 3.680059349536896, "ce": 5.543869137763977, "genuineness": 3.561077189445496, "identity": 0.4433265455067158, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.10179421678185463, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.5406249594169152, "target_raw": 3.39995082616806, "ce": 5.732011175155639, "genuineness": 3.6137959361076355, "identity": 0.46651800274848937, "wer": 0.015625, "valid_fraction": 0.95, "identity_delta": -0.07860275954008103, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.534374982190157, "target_raw": 3.391976237297058, "ce": 5.663406276702881, "genuineness": 3.5985764980316164, "identity": 0.4328148167580366, "wer": 0.011228070175438596, "valid_fraction": 0.95, "identity_delta": -0.11230594553053379, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.4642578104744952, "target_raw": 3.302511012554169, "ce": 5.7059989213943485, "genuineness": 3.617247724533081, "identity": 0.4613871800713241, "wer": 0.03014254385964912, "valid_fraction": 0.95, "identity_delta": -0.08373358221724629, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.1129882921713021, "target_raw": 2.8543125808238985, "ce": 5.849264740943909, "genuineness": 3.704798460006714, "identity": 0.4849243342876434, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.06019642800092697, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -1.0503417948266855, "target_raw": 0.09403563737869262, "ce": 6.168361663818359, "genuineness": 4.26383672952652, "identity": 0.5451207622885704, "wer": 0.0059649122807017545, "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": -1.0266113248084643, "target_raw": 0.12431426644325257, "ce": 6.188218307495117, "genuineness": 4.29804117679596, "identity": 0.5321254022419453, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.012995360046625137, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -1.0572265568979284, "target_raw": 0.08525110185146331, "ce": 6.1878211975097654, "genuineness": 4.270740389823914, "identity": 0.5377919435501098, "wer": 0.006274509803921568, "valid_fraction": 1, "identity_delta": -0.007328818738460541, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -1.1103515641970445, "target_raw": 0.017466840147972108, "ce": 6.203706550598144, "genuineness": 4.341032087802887, "identity": 0.5874357163906098, "wer": 0.01111111111111111, "valid_fraction": 1, "identity_delta": 0.04231495410203934, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.9619531219708669, "target_raw": 0.2068141806125641, "ce": 6.235000667572021, "genuineness": 4.330864872932434, "identity": 0.5942860591411591, "wer": 0.012958780350084697, "valid_fraction": 0.98, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.598515642437588, "target_raw": 3.4738157987594604, "ce": 5.779726734161377, "genuineness": 3.631211919784546, "identity": 0.4897233945131302, "wer": 0.012799736495388668, "valid_fraction": 1, "identity_delta": -0.10456266462802886, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.2637304859010976, "target_raw": 3.046650404930115, "ce": 5.945490226745606, "genuineness": 3.687194199562073, "identity": 0.5347678869962692, "wer": 0.01113683418031244, "valid_fraction": 1, "identity_delta": -0.05951817214488983, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.17757813978071, "target_raw": 2.936725277900696, "ce": 5.947078790664673, "genuineness": 3.7133653020858763, "identity": 0.5075179040431976, "wer": 0.01296207415772633, "valid_fraction": 1, "identity_delta": -0.08676815509796143, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.0861523461203879, "ci95_z": [ -0.06556643241915547, 0.24556641186484493 ], "language_deltas": { "en": -0.07451168330366816, "de": 0.391406261459612, "fr": 0.11289067321066089, "nl": 0.1281249582490568, "es": -0.127148479013722 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.0015885639190673828, "delta_genuineness": -0.02617110252380371, "delta_wer": -0.0018252399774138905, "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": -0.33478515653649027, "ci95_z": [ -0.488652320193928, -0.14453126689015433 ], "language_deltas": { "en": -0.4311523652732681, "de": 0.012499935767779635, "fr": -0.4652343177384353, "nl": -0.24921873532877695, "es": -0.5408203001097507 }, "languages_improved": 1, "clear_gain": false, "delta_ce": 0.1657634925842285, "delta_genuineness": 0.05598227977752686, "delta_wer": -0.001662902315076228, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 2.560468764408455, "ci95_z": [ 2.333203174407716, 2.786093758861127 ], "language_deltas": { "en": 2.162500052553635, "de": 2.7917969568595855, "fr": 2.8855468254754983, "nl": 2.3867187575180893, "es": 2.5757812296354663 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.45527393341064454, "delta_genuineness": -0.6996529531478882, "delta_wer": -0.00015904385469602849, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 1.0, "reason": "No new configuration passes the held-out gates; retain standard fallback and inspect audio", "fallback_standard": true }, "reference_only": true, "updated": 1793044555.2284143 }, "rows": [ { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "0001f3570b23bb5ed3d5cd3da1944fe9b3c62626fd829cdea5884f627a445a16", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84f60473d2fd8fbac9471b72171d08f8aec8253983c743da292c9ea87bfb5c17", "generation_seconds_batch": 17.501822380931117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4160823822021484, "target_z": 0.7695312430658399, "ce": 5.657250881195068, "genuineness": 3.8044304847717285, "identity": 0.582709014415741, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0001f3570b23bb5ed3d5cd3da1944fe9b3c62626fd829cdea5884f627a445a16.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "003d047a3e79b60b4e63f7d0cca31bab7b2269c86d647a466833249facf1d090", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "46744fa081a4fe828ab2cd3bab377b64183d4e56d5ebb7788c99ae61d3c271d5", "generation_seconds_batch": 22.93579800100997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627227306365967, "target_z": 1.7187498569373274, "ce": 6.09409761428833, "genuineness": 3.8295347690582275, "identity": 0.7103992104530334, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/003d047a3e79b60b4e63f7d0cca31bab7b2269c86d647a466833249facf1d090.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": "005cf87e8bfcb75b95fcef08ea93d983ad8f4fd12f1a50d3836b9f385330a15d", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "22e511d51bfc94bd869d3de8c7c50382b9d5b6259efd941126a1e3621caf9f46", "generation_seconds_batch": 21.73623162601143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.507608413696289, "target_z": 1.6250001518216117, "ce": 6.062326908111572, "genuineness": 3.823258876800537, "identity": 0.28540512919425964, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/005cf87e8bfcb75b95fcef08ea93d983ad8f4fd12f1a50d3836b9f385330a15d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "00af4fb679a4af3d78d07b90bf1f82f7e70903eff7db10aaa912e97b21997be9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f46da14e65f3644deeadbbdc2e488b1e3e84c0d14af48ab4a01b79ce8041820f", "generation_seconds_batch": 15.114476414979436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8410953283309937, "target_z": -0.46484376715292247, "ce": 6.23706579208374, "genuineness": 3.9801597595214844, "identity": 0.5203019380569458, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/00af4fb679a4af3d78d07b90bf1f82f7e70903eff7db10aaa912e97b21997be9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "00c564dbc7d57f30ac8e28839115d17c98100f86253a852b188abbf111f1bf48", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "98e4c42bb4781b9cc5dacce29e2aee847b26da69c4e4334a5b5f086a507b8db5", "generation_seconds_batch": 19.76215807988774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7898926734924316, "target_z": 1.0625000525536348, "ce": 5.387199878692627, "genuineness": 3.7040140628814697, "identity": 0.3240169584751129, "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_RANT_reg/00c564dbc7d57f30ac8e28839115d17c98100f86253a852b188abbf111f1bf48.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "00f5ae4bd3134974e5900aeec73611f61f2ede221d7e7af066bb44f3c0a3fa85", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f29c01b2a8e64e134d8c0fdc963c27ab14347715f6a48571e9b46a4939f5c938", "generation_seconds_batch": 22.5495894550113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.241637706756592, "target_z": 0.632812564962132, "ce": 6.0305562019348145, "genuineness": 3.722842216491699, "identity": 0.6774616241455078, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/00f5ae4bd3134974e5900aeec73611f61f2ede221d7e7af066bb44f3c0a3fa85.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "012e33b7939b3f043401c8bf667a6b06abb28720dd349498a7866200a3036918", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bd8754e401cf038e99888094d4b063507d0d3930b04c75789c9bc1edb419324e", "generation_seconds_batch": 22.28690435795579, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.697005271911621, "target_z": 1.7734374029217579, "ce": 5.879645347595215, "genuineness": 3.487490653991699, "identity": 0.0012309551239013672, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/012e33b7939b3f043401c8bf667a6b06abb28720dd349498a7866200a3036918.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "a8ee2bd1b2b8abef9ac89c5d31fff19791b79503fcfd4e988d5719ef0e6cf9af", "generation_seconds_batch": 17.845271227066405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06103324890136719, "target_z": -1.1718749576651275, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.8026185631752014, "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_RANT_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "025c3a41ad8440b2713a34974d338079f253aff2b644611ab8205085a0f34ab3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8719ce26708caa7c47f17edaf5528778d3ee103ada479774200a4b537aa5f2b4", "generation_seconds_batch": 16.640784753020853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8664660453796387, "target_z": 1.906250014598232, "ce": 5.7406487464904785, "genuineness": 3.3996260166168213, "identity": 0.5816217660903931, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/025c3a41ad8440b2713a34974d338079f253aff2b644611ab8205085a0f34ab3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "02b9f97e60aa91e3bcbbb7da181d01401fa0cb9f107d237edae9e097956ad1e7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2946d085d75590f8842a87889696486c383fed89de5fc4740ce2111762b1417a", "generation_seconds_batch": 24.25824624102097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6902098655700684, "target_z": 0.9843749868615913, "ce": 6.0305562019348145, "genuineness": 3.8044304847717285, "identity": 0.6174225807189941, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/02b9f97e60aa91e3bcbbb7da181d01401fa0cb9f107d237edae9e097956ad1e7.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": "031b252137ecbcda7cb92626c0a5f91d53a8417e86053a15e23f9306773b7968", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c634a6125d3767cf839b0e1076f26b4ec922108e9694f942e3c8a3aea930eb42", "generation_seconds_batch": 16.696319916984066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0690040588378906, "target_z": 1.2812498627766202, "ce": 5.728734970092773, "genuineness": 3.5690791606903076, "identity": 0.6368028521537781, "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_RANT_reg/031b252137ecbcda7cb92626c0a5f91d53a8417e86053a15e23f9306773b7968.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "03ac1f780b8f49b5d6e1b5e29a1ac1cf5f927bad1a768958d3f436e4f05b5736", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e1240c4a67504637eace482e47f0caf8b0ff56171c0ac3b8c8e27fac2bde94dd", "generation_seconds_batch": 21.80460810905788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.138782024383545, "target_z": 1.3359374087610507, "ce": 5.744620323181152, "genuineness": 3.7918784618377686, "identity": -0.01584758795797825, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/03ac1f780b8f49b5d6e1b5e29a1ac1cf5f927bad1a768958d3f436e4f05b5736.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "03f3bbb46335da9f3f6cbf7e3645137d751611fbb3dd4196ef3d6eedf62d5eff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "855caa823fc9874f95fc4511d95ee765aaf24ee31300d51f7159b24ca61590db", "generation_seconds_batch": 16.85546148591675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9283177256584167, "target_z": -0.3964843813867265, "ce": 6.046441555023193, "genuineness": 4.193545341491699, "identity": 0.6193253397941589, "hypothesis": "La porte 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_RANT_reg/03f3bbb46335da9f3f6cbf7e3645137d751611fbb3dd4196ef3d6eedf62d5eff.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": "04c70f6582fb368876336d0ea450dcb58d59eaba1ca8ec170bf9428e0f6e6713", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c658e597ebc33e9960cb947be07daa2c1cf1343985b88252623efc039d7344c", "generation_seconds_batch": 16.280907156062312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.168686866760254, "target_z": 1.3593749284686636, "ce": 5.879645347595215, "genuineness": 3.8671910762786865, "identity": 0.20944136381149292, "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_RANT_reg/04c70f6582fb368876336d0ea450dcb58d59eaba1ca8ec170bf9428e0f6e6713.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "fb68949724c125848a3d9df60b531203edb827b38e08ccc3e400e8c13431bf5c", "generation_seconds_batch": 15.23482078302186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08849072456359863, "target_z": -1.0546875459844305, "ce": 6.3641486167907715, "genuineness": 4.3881025314331055, "identity": 0.15173789858818054, "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_RANT_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "05e6f0df4e3fdd75d6bd14bbc962b74294eb645f7e568c2c842cf30930c4679a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7454b30949457b786e3163540de3e42aaa2e0ddfa0116e3c4034d24ef6bc0916", "generation_seconds_batch": 22.749278461094946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.746962070465088, "target_z": 0.24511716003707623, "ce": 5.577824115753174, "genuineness": 3.9299514293670654, "identity": 0.7154272794723511, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/05e6f0df4e3fdd75d6bd14bbc962b74294eb645f7e568c2c842cf30930c4679a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "0679b51ff791380bb0c3d792bd1de8991a145eabad103cfc84288e767ca07e10", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ed4f8497bed4f3614c2748fa5686ae1d66a2d5cd99d4f052093fb6dc2a8f2021", "generation_seconds_batch": 22.184799488983117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049067735671997, "target_z": 1.2656250364955797, "ce": 5.7128496170043945, "genuineness": 3.591045379638672, "identity": 0.4464961588382721, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0679b51ff791380bb0c3d792bd1de8991a145eabad103cfc84288e767ca07e10.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "07b00a564ad6cc495689f4ab10135e9ec3b18cd34b8a9958921eec88fe7060db", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "7b1263ab4dac114f5f4cd4d22d3da583a7d0515d1f048358e54c99ddede4a79a", "generation_seconds_batch": 27.142972182948142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4459872245788574, "target_z": 0.792968762773453, "ce": 6.0146708488464355, "genuineness": 3.6412534713745117, "identity": 0.5706812143325806, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/07b00a564ad6cc495689f4ab10135e9ec3b18cd34b8a9958921eec88fe7060db.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "07c5faf33f90aca503c58929f85a2d044d5abe5d34672f82a52c011694b719c8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "75528aabfb30e664946bce9aeef30c291e89c6e22875eb91c0a85dd08ea645d4", "generation_seconds_batch": 20.74181874305941, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6816030144691467, "target_z": -0.5898437321171659, "ce": 5.831989288330078, "genuineness": 3.9550557136535645, "identity": 0.5962262153625488, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/07c5faf33f90aca503c58929f85a2d044d5abe5d34672f82a52c011694b719c8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "07fae40e7e1767afb523691060a8d09c9822b6ba9b5a4cea25ee07cd781a46ed", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f78624ac1a8a8c708d265955a615615600eb5d3fbf9f68a6482ce130e8a1dc97", "generation_seconds_batch": 18.43499487999361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7799243927001953, "target_z": 1.0546875459844305, "ce": 5.557967185974121, "genuineness": 3.5314228534698486, "identity": 0.4531804621219635, "hypothesis": "The door is still open! I put your coat on the chair! Tell me when you are ready!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/07fae40e7e1767afb523691060a8d09c9822b6ba9b5a4cea25ee07cd781a46ed.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "0803727687434739e6ef9a63c3aaf85ef5275443f2ebb32165a2a007ba2f4db0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9bcacec0bea32258eadb1f145e9349c5390bcc2b510e388a2667f1eebf81ff6", "generation_seconds_batch": 20.373611327027902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.25043046474456787, "target_z": -1.320312489051326, "ce": 6.125868320465088, "genuineness": 4.413206577301025, "identity": 0.5288979411125183, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0803727687434739e6ef9a63c3aaf85ef5275443f2ebb32165a2a007ba2f4db0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "0858f4d48c03dbaa257a10892660aca7c0d7c688068e65a89abece0a71681836", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "aca759232ff690cacba7c927f4c9923cb19a53e2245304c803cab56b8ff3952c", "generation_seconds_batch": 20.21807613305282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.1009063720703125, "target_z": -1.203124983941945, "ce": 5.871703147888184, "genuineness": 4.118232727050781, "identity": 0.69464510679245, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0858f4d48c03dbaa257a10892660aca7c0d7c688068e65a89abece0a71681836.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "08a3688f4a7f0325ae24ac3af2b8ad863ce709ee65cee603447677d1e208c13e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3cbf8d8624f66c245c0d525af0ebe8a9dd8079440288f98a241a7b2ca3778dde", "generation_seconds_batch": 22.14719259995036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9262757301330566, "target_z": 1.953125054013458, "ce": 5.863759994506836, "genuineness": 3.6412534713745117, "identity": 0.6487376689910889, "hypothesis": "J'ai trouvé la 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_RANT_reg/08a3688f4a7f0325ae24ac3af2b8ad863ce709ee65cee603447677d1e208c13e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "09872173260220600f27e5df6aee4f125d4dac79a453205ec6598117da83ce39", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2e625f713deff1b91ebaf01db4d631863f6b98b458cb9a66f54fced34ce18a55", "generation_seconds_batch": 16.475742437993176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.407925605773926, "target_z": 1.5468750861295681, "ce": 5.597680568695068, "genuineness": 3.71028995513916, "identity": 0.4527568817138672, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/09872173260220600f27e5df6aee4f125d4dac79a453205ec6598117da83ce39.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "09e31fa92ca384516096c0effb2c54e135d12c3515c57439bf5d56b7d05a1143", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "cc65b9f0803275edf0cfb165e055cb42fee6e7cdbe086adde4e273bbf3d5259e", "generation_seconds_batch": 23.55741221399512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.251605987548828, "target_z": 0.6406250715313363, "ce": 6.125868320465088, "genuineness": 3.9048471450805664, "identity": 0.4027484655380249, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/09e31fa92ca384516096c0effb2c54e135d12c3515c57439bf5d56b7d05a1143.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": "0a3732eeb420aa9a6cc17d14b656d2487b0ee61771e5f9c1822207d4315321ef", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2e587ee909a7681adf809ef548347a3afa1e1791f27b069493521ba708f150ad", "generation_seconds_batch": 21.040634370059706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7069735527038574, "target_z": 1.7812499094909622, "ce": 5.974957466125488, "genuineness": 3.5941832065582275, "identity": 0.5045199394226074, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0a3732eeb420aa9a6cc17d14b656d2487b0ee61771e5f9c1822207d4315321ef.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "0a40a7d0c09416a809d84839cdaf21458b30dd4531ec349e06c971dc816faf31", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9c12be2db2f04c4d0fdb0ed4554d319a3a7d50c5717159008fd1f0b7c4409dc4", "generation_seconds_batch": 23.16291715798434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06855428218841553, "target_z": -1.070312465694155, "ce": 6.324435234069824, "genuineness": 4.093128204345703, "identity": 0.583987832069397, "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_RANT_reg/0a40a7d0c09416a809d84839cdaf21458b30dd4531ec349e06c971dc816faf31.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": "0aa11019006a33c5dad5209c3f07784be9b83d79d9a5ecf86428a7deb0da748c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b87fe7025a7781a5b6e0e4f81928c604bf9c9087d52b136b4ab960462d99823c", "generation_seconds_batch": 20.861759269959293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5175766944885254, "target_z": 1.632812658390816, "ce": 6.1020402908325195, "genuineness": 3.6977379322052, "identity": 0.7091426849365234, "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_RANT_reg/0aa11019006a33c5dad5209c3f07784be9b83d79d9a5ecf86428a7deb0da748c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "0aa5b46efbcfca344c0c89af3da3fe93515d42773d5014d7b0666579633790a0", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bdce9c2fa225a9c815385e4bebceb88338b0c37c48796b6ee97263e620536eea", "generation_seconds_batch": 18.01342276192736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.238014817237854, "target_z": -0.9375000408750493, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.7792433500289917, "hypothesis": "J'ai trouvé la 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_RANT_reg/0aa5b46efbcfca344c0c89af3da3fe93515d42773d5014d7b0666579633790a0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "0b8ea6bb463260453f71321638f2d85ce953ca1e4a1ed8844835cee8d6bbd825", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "e3108e34baa32809651bf19eca9aeb0bf8148265a3f49a9dd6eacb956f3b1a31", "generation_seconds_batch": 25.67660490097478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120500564575195, "target_z": 0.5312499795624753, "ce": 5.8875885009765625, "genuineness": 3.653805732727051, "identity": 0.41987335681915283, "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_RANT_reg/0b8ea6bb463260453f71321638f2d85ce953ca1e4a1ed8844835cee8d6bbd825.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "0bbe605eb25ed5eea444c6e7066d57e0490b196efb51fad63c4814c520e19c52", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90b6445c49f847c2cdbe3b1be6a36479240ecb3eaa7907d5f0c5de96245f4929", "generation_seconds_batch": 19.23910531308502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9824625253677368, "target_z": 0.4296874875915029, "ce": 6.062326908111572, "genuineness": 3.936227321624756, "identity": 0.4747544825077057, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0bbe605eb25ed5eea444c6e7066d57e0490b196efb51fad63c4814c520e19c52.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "0bcb323461abdd1462b6ffe4522d8f9d5ee8994226eab813f0816512368eac0c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ec14d0e751e8d34acbdf29fc8b3ccb64c965982b530b8b2a1c6d8cac53793722", "generation_seconds_batch": 18.297362613026053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1786551475524902, "target_z": 1.367187435037868, "ce": 6.086154937744141, "genuineness": 3.600459337234497, "identity": 0.47248023748397827, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0bcb323461abdd1462b6ffe4522d8f9d5ee8994226eab813f0816512368eac0c.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "394f2eaeb07ebea4bc52ef520ec04c55186d09011e4d72d2b00713674d0445d3", "generation_seconds_batch": 17.921695042052306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12836384773254395, "target_z": -1.023437519707613, "ce": 6.1894097328186035, "genuineness": 4.432034492492676, "identity": 0.7123674154281616, "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_RANT_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "0cbd3d53272f9e64f1f29c6d98d485fb713988ebfa41a9f7dae065d3c3beb61e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "15f7448d530b8f9b31371b03124bce96fbad2498262dc7fd1d2991c09d421c3b", "generation_seconds_batch": 23.78331406810321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.541366457939148, "target_z": 0.08398435219026265, "ce": 5.935244083404541, "genuineness": 3.923675537109375, "identity": 0.5232795476913452, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0cbd3d53272f9e64f1f29c6d98d485fb713988ebfa41a9f7dae065d3c3beb61e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "0cf6170f8899f56bc4512da19bb3b6a1490a302e3afb438d05d24f7f6fa7c5a7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "292ef43041252e2742586ea1ecb16348a07060b2b334f4ac7cac8f666779f008", "generation_seconds_batch": 23.16291715798434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6516981720924377, "target_z": -0.613281251824779, "ce": 6.300607204437256, "genuineness": 4.293961524963379, "identity": 0.4109627902507782, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0cf6170f8899f56bc4512da19bb3b6a1490a302e3afb438d05d24f7f6fa7c5a7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "0d480011cd497fa12dba9098201f70ca09869e75db05a38ea18dd8bf57657940", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d8f1152eb146e32f7ed943b9072176cad86faa302152759e2f4f59ad5819122d", "generation_seconds_batch": 22.083597937016748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45731687545776367, "target_z": -0.7656249897812377, "ce": 6.133810997009277, "genuineness": 4.356721878051758, "identity": 0.325006902217865, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0d480011cd497fa12dba9098201f70ca09869e75db05a38ea18dd8bf57657940.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "0e7bfd86ef433d5434071a4a04b0400c495309a92bfa50eb96451b929ff1f8d1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "98839ab9be40cb40bd5c3c7e973ab04f502299de1d538c2c971f5961831cf186", "generation_seconds_batch": 22.64774499705527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5619837045669556, "target_z": -0.6835937642332761, "ce": 5.895530700683594, "genuineness": 4.193545341491699, "identity": 0.5626783967018127, "hypothesis": "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_RANT_reg/0e7bfd86ef433d5434071a4a04b0400c495309a92bfa50eb96451b929ff1f8d1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "0ecb595540acd7152a4b412c8aae23ca5037cb24af4249850c1d266db23e7898", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79e1890cae1a99443923c625ac91044e73e0222c92367fe58fc1d50222e0a492", "generation_seconds_batch": 22.51990895497147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8497023582458496, "target_z": 1.109375091968861, "ce": 5.677107334136963, "genuineness": 3.587907314300537, "identity": 0.49762773513793945, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¡Podemos mirar juntos lo que hay dentro!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0ecb595540acd7152a4b412c8aae23ca5037cb24af4249850c1d266db23e7898.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "0f03c547fb3780b5e1e84358faa0585e128469b9f0daf5c60a430d45550762ef", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "94e9e2cfc44ca7e095999226c663ab98d986260878042e39ec5d2b99318afa91", "generation_seconds_batch": 23.13650703895837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1483004093170166, "target_z": -1.0078125065692043, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.6816668510437012, "hypothesis": "La porte 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_RANT_reg/0f03c547fb3780b5e1e84358faa0585e128469b9f0daf5c60a430d45550762ef.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "0f2562b7bcacf36aab3ae62b895211845e7b9fc968200117fa3d15372b8e2955", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1094773d95e5cbe5a906498d84f6f15422dc9da51754abdf2250716c0df74037", "generation_seconds_batch": 22.60545622196514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2484331130981445, "target_z": 1.4218749810222986, "ce": 5.2065043449401855, "genuineness": 3.6851859092712402, "identity": 0.5380419492721558, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/0f2562b7bcacf36aab3ae62b895211845e7b9fc968200117fa3d15372b8e2955.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "0f625360fe40f00c72e77da4d2f8c79d5dfb7f8fae16f5a24386e520b25194cf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "3063188adf7d16dc7c340cf4444ee6504d78589b707b444ca330866bc0bd7c37", "generation_seconds_batch": 24.352225829963572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6921366453170776, "target_z": 0.20214846733513644, "ce": 5.77639102935791, "genuineness": 3.6851859092712402, "identity": 0.5443578958511353, "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_RANT_reg/0f625360fe40f00c72e77da4d2f8c79d5dfb7f8fae16f5a24386e520b25194cf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "0f93b9b5a0d9f412f5df9b57f3b3fc8048fafadf49b4bac219d9f7296f5fac1e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "41dbb31d073b7d29eede35badf858cc81045c2fec17eef0bc581c4194a0b56f7", "generation_seconds_batch": 16.560774549026974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.879645347595215, "genuineness": 3.6412534713745117, "identity": 0.47780805826187134, "hypothesis": "La porte 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_RANT_reg/0f93b9b5a0d9f412f5df9b57f3b3fc8048fafadf49b4bac219d9f7296f5fac1e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "1019537c33d7275d93071a080fd44b3c33f7cfa3858201b58d84f2e46624fc80", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0cfe039f02f4d03e70157282a4cd7e87c14b946d14e18bd6979a7107a056121c", "generation_seconds_batch": 22.51990895497147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6603050231933594, "target_z": 0.9609374671539782, "ce": 5.19657564163208, "genuineness": 3.6851859092712402, "identity": 0.6928884983062744, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1019537c33d7275d93071a080fd44b3c33f7cfa3858201b58d84f2e46624fc80.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": "10a64d3402e8cc0b8eb9dd8aa14f13f4688db7fae6546e3e5e862c4762f091c1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "710f46802c63be834497ee0ee106d4de13516169f42c3bb6d5ce0503547d80da", "generation_seconds_batch": 21.744843691936694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7930653095245361, "target_z": 0.2812499562053043, "ce": 5.919358730316162, "genuineness": 3.6789097785949707, "identity": 0.3943030834197998, "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_RANT_reg/10a64d3402e8cc0b8eb9dd8aa14f13f4688db7fae6546e3e5e862c4762f091c1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "110e2a2503e9263ef027e53ffd07a82dfa106539429ee297d432ef740843f69f", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b9631149ff7a4bd0681a5273c26ed45260ad2219fcd214943016b13d4647dde1", "generation_seconds_batch": 17.74108366202563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0585860013961792, "target_z": -1.0781249722633595, "ce": 6.23706579208374, "genuineness": 4.3504462242126465, "identity": 0.34869036078453064, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/110e2a2503e9263ef027e53ffd07a82dfa106539429ee297d432ef740843f69f.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": "1155a444f36c22c88d14b5a8e5400e7765ca92dce6d112202ade9cbdb9952b42", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2050bef2667fbbdc2c90526b04798419d0c17d349e4ad29ec5ac5f0ea69bf18b", "generation_seconds_batch": 23.706111781997606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2783379554748535, "target_z": 1.4453125007299117, "ce": 5.77639102935791, "genuineness": 3.5722172260284424, "identity": 0.596066951751709, "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_RANT_reg/1155a444f36c22c88d14b5a8e5400e7765ca92dce6d112202ade9cbdb9952b42.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": "11b2bdc8c94972ebfb710c790145929f9a1dde0daeae7150ce7f2ffc66dca9af", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6edf71af1d1945685117e27efddbaba7c79ed167c6d8d1cb9b0c79f7e8fd57b", "generation_seconds_batch": 30.375837830011733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8497023582458496, "target_z": 1.109375091968861, "ce": 5.677107334136963, "genuineness": 3.613011360168457, "identity": 0.6180288791656494, "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_RANT_reg/11b2bdc8c94972ebfb710c790145929f9a1dde0daeae7150ce7f2ffc66dca9af.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "11e45f34fadc610e04fcf94d85485f8df2a906c3be2f415304e754ab9cfd169e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f2397a840659b2b850769630d80af8797d283cbe9679c016b01ad3542a91f12", "generation_seconds_batch": 22.749278461094946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7369937896728516, "target_z": 0.23730465346787188, "ce": 6.260893821716309, "genuineness": 4.262581825256348, "identity": 0.37350159883499146, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/11e45f34fadc610e04fcf94d85485f8df2a906c3be2f415304e754ab9cfd169e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "121c86b870b4ded03ea41e038ffdda7b0e1687ead3fc763516d68fa9b20ef380", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e539f71fa59511aec808f2520cbd2af9333e152d9b97f095d4b818b7358e48f1", "generation_seconds_batch": 17.553768503014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9382860064506531, "target_z": -0.3886718748175221, "ce": 6.070269584655762, "genuineness": 3.8420867919921875, "identity": 0.3858189582824707, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/121c86b870b4ded03ea41e038ffdda7b0e1687ead3fc763516d68fa9b20ef380.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "1222a501fa00fc5bd2335ea43485a1518546dc493057d5bf7fbdd26d7a99eba7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ec89dea816b3dea33b6bd9118cc6f38ba2e27196b8d06fd87a6c2a628f152474", "generation_seconds_batch": 24.814282287028618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8535556793212891, "target_z": -0.45507813394141705, "ce": 5.871703147888184, "genuineness": 3.9927117824554443, "identity": 0.5984113216400146, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1222a501fa00fc5bd2335ea43485a1518546dc493057d5bf7fbdd26d7a99eba7.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "3c4bf7abd5989f2c48713cff7578acce5efd0ee319431d6b73ab52acc92cb17f", "generation_seconds_batch": 17.854703341959976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04861772060394287, "target_z": -1.0859374788325638, "ce": 6.356205940246582, "genuineness": 4.243753433227539, "identity": 0.5338790416717529, "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_RANT_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.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": "12f6652cfbb70b6bea23fbfecf522bc5879a59d737a483239649aa3faa1ca3e6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a8ff44946027c6f534f4101566d6777371c71838b39431eeeab74f8a4e54ed90", "generation_seconds_batch": 26.37386155000422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.265196800231934, "target_z": 2.2187499036516694, "ce": 6.054384231567383, "genuineness": 3.591045379638672, "identity": 0.3355036973953247, "hypothesis": "Die Straße führ hinter der Brücke nach links. Geh weiter, bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "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_RANT_reg/12f6652cfbb70b6bea23fbfecf522bc5879a59d737a483239649aa3faa1ca3e6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "1315da974ed99bc5af01615f27ecba874dbb07d3cd125593a9e6704d53ba4f26", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "666592b32fde58490c375fa5655f973ccad29bcafd1b8c237dc8e78740eb51a1", "generation_seconds_batch": 21.235691434005275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0005877017974854, "target_z": -0.339843755474337, "ce": 5.927301406860352, "genuineness": 3.5063188076019287, "identity": 0.33235594630241394, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1315da974ed99bc5af01615f27ecba874dbb07d3cd125593a9e6704d53ba4f26.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "1365808d2729b9e7c7ead20435b3dc10cd586afd94728f109412e9db70e73cdf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9c8cb1ada7ad1be3bebe60d928ff6a2eef31ab193390a4ff2770c9d1b2dc822f", "generation_seconds_batch": 20.80338961700909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7667832374572754, "target_z": 1.8281249489061884, "ce": 5.8875885009765625, "genuineness": 3.7165660858154297, "identity": 0.547746479511261, "hypothesis": "J'ai trouvé la 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_RANT_reg/1365808d2729b9e7c7ead20435b3dc10cd586afd94728f109412e9db70e73cdf.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": "137090e0a3e2b40c902a92344a8d8c4e9364f47c132595221bc31816fb3531e6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "14046fa51ae4564de52010501488bb91d3e87545e6f7dd8fa7ac65ee3791439e", "generation_seconds_batch": 16.240466531948186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4677352905273438, "target_z": 1.5937501255447943, "ce": 5.720792293548584, "genuineness": 3.6663577556610107, "identity": 0.36627596616744995, "hypothesis": "Die Straße führt hinter der Brücke nach links! Geh weiter bis zu dem kleinen Holztor! Ich warte auf der anderen Seite!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/137090e0a3e2b40c902a92344a8d8c4e9364f47c132595221bc31816fb3531e6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "13ed1c3cf8e5b2ba58f3f034de87c9416b454d010f4b93d99b6cc4f53deba47e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e287932d6055d0b6646f06123ea4319b8cba0293dfddb8898ab15f096b56a85d", "generation_seconds_batch": 15.979287159978412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1886234283447266, "target_z": 1.3749999416070724, "ce": 5.589737892150879, "genuineness": 3.785602569580078, "identity": 0.24939404428005219, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/13ed1c3cf8e5b2ba58f3f034de87c9416b454d010f4b93d99b6cc4f53deba47e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "143d150d7c0e3e82193bdd399141ab8a5fa473b3214254a5612f0a965188d68f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ddc42580efeb05a001d6bbdc1d79334e78c4619d0cde57a9f1b293532975a77f", "generation_seconds_batch": 16.449964698054828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0255084037780762, "target_z": -0.3203124890513261, "ce": 6.070269584655762, "genuineness": 3.7542223930358887, "identity": 0.25175362825393677, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/143d150d7c0e3e82193bdd399141ab8a5fa473b3214254a5612f0a965188d68f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "144db5d40a1093a300a49f64e9b3575954b0818d68d80aad9e66a46f69fc1876", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0170b1c1d5220b47e70b36dfe254a5c3789a39f8b0ab6c100ded77cf50b82ed8", "generation_seconds_batch": 18.96602002205327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8497023582458496, "target_z": 1.109375091968861, "ce": 5.935244083404541, "genuineness": 3.603597402572632, "identity": 0.6083317399024963, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/144db5d40a1093a300a49f64e9b3575954b0818d68d80aad9e66a46f69fc1876.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "1520bb506fa77cb1f061b3aed508e170e39e556a1ec2c99377c9656568f4f89a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "42b992055028cbcc2e3fd88e2d5f261d2dfceebdbd666cf1d4a8d5daa501e72b", "generation_seconds_batch": 22.184475932037458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3578879833221436, "target_z": -0.059814445027543246, "ce": 5.95112943649292, "genuineness": 4.024092197418213, "identity": 0.526989758014679, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1520bb506fa77cb1f061b3aed508e170e39e556a1ec2c99377c9656568f4f89a.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "b9aaa481800234869ba134a3763763c9ce12e8024289cda1d9700de7b7284888", "generation_seconds_batch": 14.721436911961064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.021160244941711426, "target_z": -1.1406250248169942, "ce": 6.308549880981445, "genuineness": 4.469690799713135, "identity": 0.5585612654685974, "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_RANT_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "1559e15b8fa49fe051c15fb7c3f69afd5da202d40263f699366b2b645d26539d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "04e9dc8b1eaa7f9543188428531b573a4eb3c3b8710bcbfa3174ae56bc92b534", "generation_seconds_batch": 24.081040455959737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0696849822998047, "target_z": 0.498046920072041, "ce": 5.816103935241699, "genuineness": 3.9299514293670654, "identity": 0.6601093411445618, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1559e15b8fa49fe051c15fb7c3f69afd5da202d40263f699366b2b645d26539d.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": "1561c9d7308dd498e4ea8bc939695eb7eb33ee86f0d6b36def78a05dfafcde63", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f636c048cea62df777819888328b55405459a20756b7b9071d7a62b0a000df8d", "generation_seconds_batch": 16.07907515298575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.138782024383545, "target_z": 1.3359374087610507, "ce": 5.530168056488037, "genuineness": 3.785602569580078, "identity": 0.605187177658081, "hypothesis": "La carretera gira a la izquierda tras el puente. ¡Sigue hasta la pequeña puerta de madera! ¡Te esperaré al otro lado!", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1561c9d7308dd498e4ea8bc939695eb7eb33ee86f0d6b36def78a05dfafcde63.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "161365817f6639135f763b3e49b3d71c614430db6d566a49ac2c009429f22bc4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "ce5b6a586986b05985ed1e4f1a073e2eb12d5485de53fb8bde6c8ea08e0641a0", "generation_seconds_batch": 28.15482666902244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8564977645874023, "target_z": 1.8984375080290274, "ce": 5.4944257736206055, "genuineness": 3.5314228534698486, "identity": 0.6120989918708801, "hypothesis": "¡La puerta sigue abierta! He dejado tu abrigo sobre la silla. Dime cuándo estás listo!", "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_RANT_reg/161365817f6639135f763b3e49b3d71c614430db6d566a49ac2c009429f22bc4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "1625e3d1f37657ccb777989dedaaaf6b5877ff8279ebc85c91ff2558377bbf20", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bab9bc85c0998efb110ce14f453b14d24f6f3a54b29239e1e5824498a237d403", "generation_seconds_batch": 23.992605827050284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8759843111038208, "target_z": -0.43749999416070723, "ce": 6.1020402908325195, "genuineness": 4.268857479095459, "identity": 0.5472287535667419, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1625e3d1f37657ccb777989dedaaaf6b5877ff8279ebc85c91ff2558377bbf20.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "1674d768caca51c679221a702007b0eb1b76045a07c425846cdf02bcc8c93840", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f90a00953dfe9212237fa6fb0a439462822b6d0dad6a27b8d8aa507c1a8aea20", "generation_seconds_batch": 15.423976648016833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6410491466522217, "target_z": 0.16210932445362206, "ce": 5.577824115753174, "genuineness": 3.8295347690582275, "identity": 0.5233463048934937, "hypothesis": "La porte 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_RANT_reg/1674d768caca51c679221a702007b0eb1b76045a07c425846cdf02bcc8c93840.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "167657f580af285c94b3655070190a5eb445b08a6d03f42305082d471d20edaa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "51d89b5902993f7fd4eb3f1bdeeb83e7838d4788fda33a95cd06df97700cf6de", "generation_seconds_batch": 23.694552871980704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3263678550720215, "target_z": 0.6992186839430007, "ce": 6.0146708488464355, "genuineness": 3.4843525886535645, "identity": 0.35226544737815857, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/167657f580af285c94b3655070190a5eb445b08a6d03f42305082d471d20edaa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "168a2585a707d3fca4b06ce82b125b7859ecfbe9e5387ad1f54fb3580cdc31f8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f9cca6796c7abd90ac4cd79657a0b6d87b215737cd2eef1a7f89d6801756bfb9", "generation_seconds_batch": 22.83529062604066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2366535663604736, "target_z": 0.6289063116775298, "ce": 6.086154937744141, "genuineness": 4.068024158477783, "identity": 0.6241738796234131, "hypothesis": "La porte 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_RANT_reg/168a2585a707d3fca4b06ce82b125b7859ecfbe9e5387ad1f54fb3580cdc31f8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "1702cc92ebae57c5c711dcc08be67442f8b125c26bb438ee28b354379d0ee297", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "e3f0b249cb6bb74215ae12dcb603ad7ed44363416831346c8ca52eb306e8f865", "generation_seconds_batch": 22.821869064006023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.5659098625183105, "genuineness": 3.8546390533447266, "identity": 0.15431338548660278, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt! Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1702cc92ebae57c5c711dcc08be67442f8b125c26bb438ee28b354379d0ee297.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "17299a26f280c2e8b16027f173ab192c639f3e1cae3d37395a75ac6b1f3f96e4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48d75ddbdb53e6ffcb9aed4f1973a78d6534b6cbfd180b5e94d9aaf321ecba6c", "generation_seconds_batch": 20.677993583027273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8265929222106934, "target_z": 1.8749999883214146, "ce": 4.65250301361084, "genuineness": 3.578493118286133, "identity": 0.5822906494140625, "hypothesis": "¡La puerta sigue abierta! He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/17299a26f280c2e8b16027f173ab192c639f3e1cae3d37395a75ac6b1f3f96e4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "1783b1c48d43f7af12c7c7937ba9a8fd0b72a576256578d9b572828cd11afb46", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e41ec88013b8043a67dc77328072aed3d69e6f480de3b3295eb5e18a2bc0fd39", "generation_seconds_batch": 19.67135965800844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9163074493408203, "target_z": 1.9453125474442536, "ce": 5.33954381942749, "genuineness": 3.205068826675415, "identity": 0.31184542179107666, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1783b1c48d43f7af12c7c7937ba9a8fd0b72a576256578d9b572828cd11afb46.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "17cdb7c5e75c340c87080e643744217172ee8355995b38f2da052d269e5c0994", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d6e29da8484aa83822d55485b1d422f62bba9b478613a8d5a12032d2d51e5616", "generation_seconds_batch": 22.41062505193986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9457622170448303, "target_z": -0.38281249489061886, "ce": 6.125868320465088, "genuineness": 4.375550270080566, "identity": 0.7146117687225342, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/17cdb7c5e75c340c87080e643744217172ee8355995b38f2da052d269e5c0994.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": "18ee5363d0bfc16b9c6c947fad024429a6cdd2ec9652f650e617db27c0740bbb", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c17390fea0d47c31c7f1af7d1891589700fc63c2ea53c5c2af9b085dee9776d", "generation_seconds_batch": 21.067508641048335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4876718521118164, "target_z": 1.6093751386832031, "ce": 6.078212261199951, "genuineness": 3.4466962814331055, "identity": 0.5029383897781372, "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_RANT_reg/18ee5363d0bfc16b9c6c947fad024429a6cdd2ec9652f650e617db27c0740bbb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "190941e483851e32b9a5f355f26fa0c5a911eda66a196df3425eb7ed6b29c040", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5c53c7176486fcd76c01eb06c97f1d57c85b4aaf5cc33a035169799369d0a1c7", "generation_seconds_batch": 14.957987376954406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9656987190246582, "target_z": -0.3671875284665522, "ce": 5.959072113037109, "genuineness": 4.055472373962402, "identity": 0.662734866142273, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/190941e483851e32b9a5f355f26fa0c5a911eda66a196df3425eb7ed6b29c040.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "1972857d7c8e7ae25ee97dd720120600d3dfa3e84e4bc46cf606a32f8e7b3ee7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "dc4039cb92cdfca5fe19bed1c30107d6c0dab56ba33d781f29c05779345f7f62", "generation_seconds_batch": 23.479194947052747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8397340774536133, "target_z": 1.1015625853996567, "ce": 6.125868320465088, "genuineness": 3.6977379322052, "identity": 0.7721738815307617, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1972857d7c8e7ae25ee97dd720120600d3dfa3e84e4bc46cf606a32f8e7b3ee7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "19cba3a7a109afa96e52ab28524558b3e21e96a0481a146d7427fa8ad1c8b3ff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1e4047be2d067bcc2c572cd70d7bf6833796711aeb9b46c3402353bb985ca227", "generation_seconds_batch": 23.082436795928515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6702733039855957, "target_z": 0.9687499737231826, "ce": 6.221180438995361, "genuineness": 3.9927117824554443, "identity": 0.522667407989502, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/19cba3a7a109afa96e52ab28524558b3e21e96a0481a146d7427fa8ad1c8b3ff.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": "1a10479879c1a2a59a3917da02f8103c60caf69ab4bea26f5147fbf410b0f574", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f88d693b984c03b56b78cccc58771062b708e593345c75e99187408cee386a4a", "generation_seconds_batch": 21.12603548099287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.019162893295288, "target_z": 1.2421875167879668, "ce": 6.054384231567383, "genuineness": 3.8922953605651855, "identity": 0.4894862771034241, "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_RANT_reg/1a10479879c1a2a59a3917da02f8103c60caf69ab4bea26f5147fbf410b0f574.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "1ae27e1695f45efdeca62da73aa945f5507c74640413512a74d2469759d3dbb4", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3537c27d5041a5278449efa2d0ad3009cd3e4c92c8902abf96a513a1d89d0e4b", "generation_seconds_batch": 17.173635447979905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.935793936252594, "target_z": -0.3906250014598232, "ce": 5.808161735534668, "genuineness": 3.9676074981689453, "identity": 0.5132454037666321, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1ae27e1695f45efdeca62da73aa945f5507c74640413512a74d2469759d3dbb4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "1b0ad20c9e68df7e5239dbcf97009a4d14ff8e7366b0af504619e040ad041f60", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cdfc4bc19639a17dfebcc6eb686ce1eb9e9d920c6858c525434646586c36bda", "generation_seconds_batch": 22.422926418017596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34268176555633545, "target_z": -0.8554687218984036, "ce": 5.919358730316162, "genuineness": 4.20609712600708, "identity": 0.4611353278160095, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1b0ad20c9e68df7e5239dbcf97009a4d14ff8e7366b0af504619e040ad041f60.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "1b594cbc29f0212880b03e3288ddd2aeb5a484e22396130a4eee66a10860a7f8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "851b78fb92ada0e741086aeff881f9968717a9f44ecf1d659fc0d0fb1b1a25d1", "generation_seconds_batch": 22.966856123995967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4092868566513062, "target_z": -0.019531266423010885, "ce": 6.221180438995361, "genuineness": 3.9173994064331055, "identity": 0.5911122560501099, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1b594cbc29f0212880b03e3288ddd2aeb5a484e22396130a4eee66a10860a7f8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "1bc56439f2a381e838ce219d6ae90a4d876ce42283d5d4ad90cea91032739545", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "98855988637f3bcbe9bbb80fa754a96cce1c307c2d3b25c34cb1f7310046b0e9", "generation_seconds_batch": 21.57553256303072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2928403615951538, "target_z": -0.8945312547444254, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.8054306507110596, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1bc56439f2a381e838ce219d6ae90a4d876ce42283d5d4ad90cea91032739545.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "1c02f8cc99555f4ed7f63c0eebcfbb01e4c9927149136ea088c98ee5fe263a88", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "faf28738a88eba747abb850f3c95ce4ad29cb8385bbbd79d666bb20409ef9640", "generation_seconds_batch": 17.361394897918217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5569995641708374, "target_z": -0.6875000175178783, "ce": 5.534139156341553, "genuineness": 4.18099308013916, "identity": 0.39253753423690796, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1c02f8cc99555f4ed7f63c0eebcfbb01e4c9927149136ea088c98ee5fe263a88.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "1c34d9071d53b6802fbd16dad4e8c0abfbc7e24dc91ad81cfc09d1ae43f21802", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d598ce2d8349e97fd34ee2667cf18ea81a218f37b91561ea06fb37974f99dfd9", "generation_seconds_batch": 23.24568514805287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5705904960632324, "target_z": 0.890624908031139, "ce": 5.80021858215332, "genuineness": 4.036643981933594, "identity": 0.6772930026054382, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1c34d9071d53b6802fbd16dad4e8c0abfbc7e24dc91ad81cfc09d1ae43f21802.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "1ca00bb51c014a0dbbfab9492c6243807d1edf48838d53959d2dcfb4ffaa64b1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "26fe206f1eed81d140519992eba8f6846b82e520edfe536b835302f1a44b7ed4", "generation_seconds_batch": 15.139082290930673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0603973865509033, "target_z": -0.29296871605911085, "ce": 5.839932441711426, "genuineness": 4.049196243286133, "identity": 0.5829699635505676, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1ca00bb51c014a0dbbfab9492c6243807d1edf48838d53959d2dcfb4ffaa64b1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "1d04399f8b365ac93c3952de9160b6d5701d9329ddaa1de347b37afae7ff5b6b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "24a1e848959c90201d6f67c195cbcfaa11f6bab3670368f8e8c045674f87ffea", "generation_seconds_batch": 17.91086249309592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.960033893585205, "target_z": 0.4121093478107931, "ce": 5.871703147888184, "genuineness": 3.9299514293670654, "identity": 0.6600767970085144, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la casa.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1d04399f8b365ac93c3952de9160b6d5701d9329ddaa1de347b37afae7ff5b6b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "1d2f1e5e9df6399c19193b7642afba629aa580145b268ac48db098e32b502a39", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d3505ed8e7ccaa264e1d3654aeccb0b1bb8455132f147102597ea26c8e9699aa", "generation_seconds_batch": 20.836596430046484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3968265056610107, "target_z": -0.02929689963451633, "ce": 6.09409761428833, "genuineness": 4.15588903427124, "identity": 0.5413661003112793, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1d2f1e5e9df6399c19193b7642afba629aa580145b268ac48db098e32b502a39.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "1d4b77b070cf8d1a50f9ccc4c57fa4f13b2f996c4ca23b60ba07b8f2168ec0ac", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "8aea65a0e9a3a79dd459fcffb6677df94a327d1fb423170f99dcada51bc3c2a5", "generation_seconds_batch": 24.767288375995122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4872216582298279, "target_z": -0.7421875167879667, "ce": 6.316492557525635, "genuineness": 4.6956281661987305, "identity": 0.7134630680084229, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1d4b77b070cf8d1a50f9ccc4c57fa4f13b2f996c4ca23b60ba07b8f2168ec0ac.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "ff807f8a3b70ef934c83f0f4abac89cde4cf33aa84fe5d5c3573b6a80ff07db3", "generation_seconds_batch": 16.607420085929334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09845900535583496, "target_z": -1.0468750394152262, "ce": 6.300607204437256, "genuineness": 4.457139015197754, "identity": 0.7453919649124146, "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_RANT_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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "e922faaf48647423fc654764836956565a9e62f6f01f92b71d263047a22f0d4e", "generation_seconds_batch": 14.689690873026848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2629355192184448, "target_z": -0.9179687744520384, "ce": 6.324435234069824, "genuineness": 4.381826400756836, "identity": 0.44455933570861816, "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_RANT_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "1db642f86ea1462ba810cb41cbdc074f84c999d148b59e964632bf2df29fe59a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5248824e934d436fffa630c5b9d07930446a860864999c8973a5609c890615b3", "generation_seconds_batch": 21.537232307018712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.447798728942871, "target_z": 1.5781251124063855, "ce": 5.967014789581299, "genuineness": 3.4812145233154297, "identity": 0.5821735858917236, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1db642f86ea1462ba810cb41cbdc074f84c999d148b59e964632bf2df29fe59a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "1ddc9ddfe0f1f2443c12cf8881d67e267549cb421bfbb25d99361d864faaf11c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "84e22698067820c392f50b48de8d851dd1b9b1795c79bc78dd81e1ebb0fcb516", "generation_seconds_batch": 24.44391175196506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5275449752807617, "target_z": 1.6406251649600205, "ce": 6.0305562019348145, "genuineness": 3.7667744159698486, "identity": 0.3687921166419983, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1ddc9ddfe0f1f2443c12cf8881d67e267549cb421bfbb25d99361d864faaf11c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "1e4e5d8b3c863d9aa1a106e5cd18f97e413841eb429d56cb022bd30942284a37", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d4a41312d2c0aa8dc5400306eccd4ad456e5581416a4ad2c502fffe5a95a24fd", "generation_seconds_batch": 17.74108366202563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.267919659614563, "target_z": -0.9140625211674362, "ce": 6.340320587158203, "genuineness": 4.394378662109375, "identity": 0.46882563829421997, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1e4e5d8b3c863d9aa1a106e5cd18f97e413841eb429d56cb022bd30942284a37.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": "1e68c942a2b3acae35f76aa1a96e5cde9a3007ecf09daf63d8678da2fb7a9aeb", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7475e314f923081cdedf544981879e94c74a0090aa92b382b134000a888a84b9", "generation_seconds_batch": 30.582528559025377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.28830623626709, "target_z": 1.453125007299116, "ce": 6.062326908111572, "genuineness": 3.5220088958740234, "identity": 0.5621604919433594, "hypothesis": "Niemand heeft het 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_RANT_reg/1e68c942a2b3acae35f76aa1a96e5cde9a3007ecf09daf63d8678da2fb7a9aeb.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": "1e75980c23011b947cc4ec5dfaa407e8572b0ba63fae1b325e50842aaa6005c0", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29bafe9238d94e0321b7f4dc649e581f170180bce0c36e6716c4c83ce4543401", "generation_seconds_batch": 22.436992136063054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1786551475524902, "target_z": 1.367187435037868, "ce": 5.998785495758057, "genuineness": 3.4812145233154297, "identity": 0.6149168610572815, "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_RANT_reg/1e75980c23011b947cc4ec5dfaa407e8572b0ba63fae1b325e50842aaa6005c0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "1e9cd590019421b9121e7987790a353917620b0acdcef0639e8f9c496b6ca422", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "66e71cbcc0258c58018bb5b158b1dccae21bc20d57a2c9859932e6c1336ce8f6", "generation_seconds_batch": 22.41117427602876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.949384927749634, "target_z": 1.1874999708035363, "ce": 5.998785495758057, "genuineness": 3.9173994064331055, "identity": 0.5746733546257019, "hypothesis": "La porte 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_RANT_reg/1e9cd590019421b9121e7987790a353917620b0acdcef0639e8f9c496b6ca422.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "c3e9dfa8b19719b76422434377d2040d60b86e4a62f0e145ef426acf4f9423fd", "generation_seconds_batch": 16.607420085929334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2031259536743164, "target_z": -0.9648437204385804, "ce": 6.165581703186035, "genuineness": 4.545003414154053, "identity": 0.7323704361915588, "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_RANT_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "1ebb3ccbbe19f195de58c5508f46ac208a5936a974eb53ecfe6eea2e965130c8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2c88dd69f725620a0ba248827d880f74500973f0d4f7443971e065e50ba44a5f", "generation_seconds_batch": 22.391950004035607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.700935363769531, "genuineness": 3.7291181087493896, "identity": 0.5018794536590576, "hypothesis": "Son pop sonde ! J'ai trouvé la petite boîte ! Elle était près de la vieille fenêtre ! Nous pouvons regarder ensemble ce qu'elle contient !", "wer": 0.15789473684210525, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1ebb3ccbbe19f195de58c5508f46ac208a5936a974eb53ecfe6eea2e965130c8.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": "1ec581b8482eeda5ae55f2270ed2d48b96c2bcd12d4bb1f680f715f0ab4c8fc4", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "391ab97e2a900afd28785977a1cb61b26449a172bab7e30fc4ecd05070c576bc", "generation_seconds_batch": 30.235761043033563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.019162893295288, "target_z": 1.2421875167879668, "ce": 5.732706069946289, "genuineness": 3.540836811065674, "identity": 0.5260658264160156, "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_RANT_reg/1ec581b8482eeda5ae55f2270ed2d48b96c2bcd12d4bb1f680f715f0ab4c8fc4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "1ee092b9a68c2a118004be5dcf0f4a54549119c2f8581c75768b0cf3a9f85d39", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "82086de14b745b86c2e2e5c364935e2b3dc1cabd5e881c38badc0ed14779f306", "generation_seconds_batch": 22.283399573992938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.238014817237854, "target_z": -0.9375000408750493, "ce": 6.300607204437256, "genuineness": 4.105680465698242, "identity": 0.5832226872444153, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1ee092b9a68c2a118004be5dcf0f4a54549119c2f8581c75768b0cf3a9f85d39.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": "1effb34bc6a40c871c23f06abda3879abb6b36d2706ba853945cbe53f9fd6ff2", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b86c46acd0993d968ceebbca5ca05787e74157c98723af813e9e2e3a26279530", "generation_seconds_batch": 21.8379767970182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.859670639038086, "target_z": 1.1171875985380653, "ce": 6.070269584655762, "genuineness": 3.603597402572632, "identity": 0.490097314119339, "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_RANT_reg/1effb34bc6a40c871c23f06abda3879abb6b36d2706ba853945cbe53f9fd6ff2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "1f1db85e137daa37dd6f15ef60d1c694df397464da7fa82a059dab22141cc204", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9438b632a335286d2959072fddcec13c7a376e13f5a05167961d2cba9f99d91d", "generation_seconds_batch": 17.44220868300181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4559555053710938, "target_z": 0.8007812693426573, "ce": 6.213237762451172, "genuineness": 4.193545341491699, "identity": 0.5507872700691223, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1f1db85e137daa37dd6f15ef60d1c694df397464da7fa82a059dab22141cc204.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "1ffc4d8c8d2ec032956b2fb22264fa58acd2b23a08fbefa64db515aa1f8be42e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cf38059cce5fcb5509d878a6fefdee793532ae33903a5608a73ba1c9c1591d4b", "generation_seconds_batch": 20.993413344956934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.637195587158203, "target_z": 1.7265623635065317, "ce": 5.7247633934021, "genuineness": 3.487490653991699, "identity": 0.48075777292251587, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/1ffc4d8c8d2ec032956b2fb22264fa58acd2b23a08fbefa64db515aa1f8be42e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "1fffa794b1a575a4e29e64e2e2fd968fdd378fa1a76490f5e41ca6579ee8af16", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b4fc89557030aa8770e7d9ee3c111819d440080498831e9e07e7242ebb980f2b", "generation_seconds_batch": 16.78206684801262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8365612030029297, "target_z": 1.8828124948906189, "ce": 5.704906940460205, "genuineness": 3.4812145233154297, "identity": 0.7907130718231201, "hypothesis": "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_RANT_reg/1fffa794b1a575a4e29e64e2e2fd968fdd378fa1a76490f5e41ca6579ee8af16.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "201ffdc851510d3611c94d3b29b64bd4d90b09a2d53f061fca0c5d30b50deef7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a900288a8a6f381349ffcdbb77c105c94cfedd3458b16c445574721c9a4e5590", "generation_seconds_batch": 20.454556047101505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6616664528846741, "target_z": -0.6054687452555746, "ce": 6.133810997009277, "genuineness": 3.7667744159698486, "identity": 0.5119822025299072, "hypothesis": "The 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_RANT_reg/201ffdc851510d3611c94d3b29b64bd4d90b09a2d53f061fca0c5d30b50deef7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "20397e8e101589cdf91be9fd199c115842aa218b7fb167879f95fc3fdcf4bbdb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6e94702420d037273ada0fcf2b8dced8132915bfaefe89027d9d03bdbfa1c35e", "generation_seconds_batch": 18.295853212941438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.919480323791504, "target_z": 1.1640626379532915, "ce": 5.4666266441345215, "genuineness": 3.810706615447998, "identity": 0.7669548988342285, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/20397e8e101589cdf91be9fd199c115842aa218b7fb167879f95fc3fdcf4bbdb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "206e8536902ed90543e400f97c3912f23769b99d202fc0a45e0a5a3d77fcb28d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9d9a8a3d19823ce5ab9f67eb780f5efe24e6b9b76ffdf1ccc91ebd5a39d83ee1", "generation_seconds_batch": 22.854182895971462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.525733470916748, "target_z": 0.8554688153270877, "ce": 6.09409761428833, "genuineness": 3.5565271377563477, "identity": 0.6090412139892578, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/206e8536902ed90543e400f97c3912f23769b99d202fc0a45e0a5a3d77fcb28d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "20a33ccf68a4fb585507e5eebbf8e77241ed1524020b90870e119d86073dba28", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e2baa60b92a91f35476aeda710d41eee9a6af5bc7c5e61df66c6cd82926ae49f", "generation_seconds_batch": 22.422926418017596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5071582198143005, "target_z": -0.726562503649558, "ce": 5.967014789581299, "genuineness": 4.444586753845215, "identity": 0.4535251557826996, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/20a33ccf68a4fb585507e5eebbf8e77241ed1524020b90870e119d86073dba28.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "20fe5831919cbe030cd3702899f2f888905d722876a481de4bdbc5444fead218", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9625c15abb93d91640b1c79cc5eecf3528dc563c11ebaecb0f3cf18d1a5a36c5", "generation_seconds_batch": 17.553768503014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.46230101585388184, "target_z": -0.7617187364966355, "ce": 6.252951145172119, "genuineness": 3.9801597595214844, "identity": 0.5156363844871521, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/20fe5831919cbe030cd3702899f2f888905d722876a481de4bdbc5444fead218.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "216cfffe702418ba80f522a6719bdf4cb8633c91767ecfba40d561dac581b4f3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a19d9c6fc26536917d49654efe2a5e445f16ce7a83e944d257f791b164f55b1", "generation_seconds_batch": 19.76215807988774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.919480323791504, "target_z": 1.1640626379532915, "ce": 5.637393951416016, "genuineness": 3.6098732948303223, "identity": 0.256221741437912, "hypothesis": "I found the little box! It was beside the old window! We can look inside together!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/216cfffe702418ba80f522a6719bdf4cb8633c91767ecfba40d561dac581b4f3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "2177fa42a1324442427ba83f9f28d4296e0bcb6db02ae63ba025060f56154b23", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "55c93f326593558e6c329d1554e19cf5f9bf220ab5f526f6a4b4292b14c0c6e7", "generation_seconds_batch": 21.02991037000902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19315767288208008, "target_z": -0.9726562270077848, "ce": 6.316492557525635, "genuineness": 4.137060642242432, "identity": 0.5734387636184692, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2177fa42a1324442427ba83f9f28d4296e0bcb6db02ae63ba025060f56154b23.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "2196225c541b1ff8eb1b05f2151566d2a31399cb63ff4e21681e5ef1c6fc984e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c5d277d28130bae64510e68c5eb4c712c45f2124fc1384c539212bb1f32db6ed", "generation_seconds_batch": 16.769868962932378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7599878311157227, "target_z": 1.0390625328460217, "ce": 5.5539960861206055, "genuineness": 3.6914620399475098, "identity": 0.3354370594024658, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay detrás.", "wer": 0.058823529411764705, "duration_error_s": 0.120000000000001, "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_RANT_reg/2196225c541b1ff8eb1b05f2151566d2a31399cb63ff4e21681e5ef1c6fc984e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "21b7737ea0a6a0714d2e5c906a5df6a6a2a390f16fd26ca98bd2d8d5cd1ba0dc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c3f376df9c087b79840f8d9545a74d496abfc63b599db025eccb874156121171", "generation_seconds_batch": 24.072464816039428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1188454627990723, "target_z": 1.320312395622642, "ce": 5.804190158843994, "genuineness": 3.314899444580078, "identity": 0.4588262438774109, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/21b7737ea0a6a0714d2e5c906a5df6a6a2a390f16fd26ca98bd2d8d5cd1ba0dc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "222c4d057187e1b3e076aa3ebffa4a7b861c86348e8b2939d811fe10dcc723f3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c8181303538bd2ee822e7f6c8ecb6611585bd7af6ca1793d1e6b18d5ebad559a", "generation_seconds_batch": 18.633861989015713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7898926734924316, "target_z": 1.0625000525536348, "ce": 5.879645347595215, "genuineness": 3.873466968536377, "identity": 0.5990338921546936, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/222c4d057187e1b3e076aa3ebffa4a7b861c86348e8b2939d811fe10dcc723f3.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": "2266c58aa438dd2dff334c20db0bd971f9813bddc6c1acc4c412625d29b1905b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3b3386fb70e698925436b54bdd1af3b6b777b94039f0249d16020360b86d131", "generation_seconds_batch": 30.582528559025377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.378020763397217, "target_z": 1.523437566421955, "ce": 5.847874641418457, "genuineness": 3.515732765197754, "identity": 0.3929762840270996, "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_RANT_reg/2266c58aa438dd2dff334c20db0bd971f9813bddc6c1acc4c412625d29b1905b.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": "233964098afcf16ac1436eebc54bc618e5bc3751c7782139679421f69daf8d27", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ec3f870333ac500f79cebe961bcefc6f6a55ffbf2c289eacc80979185709288", "generation_seconds_batch": 15.53976936894469, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8664660453796387, "target_z": 1.906250014598232, "ce": 6.0146708488464355, "genuineness": 3.5220088958740234, "identity": 0.6647369265556335, "hypothesis": "Je bericht kwam dat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten!", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/233964098afcf16ac1436eebc54bc618e5bc3751c7782139679421f69daf8d27.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "249f79c51f131389192892eac1b4916e648ec02670bd28b5773853481f1b53de", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "abd0d7bfac4fd9d4f4ee8cded238cf54363d573837adf9988505a8d8a8ed7cd3", "generation_seconds_batch": 20.21807613305282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.11087465286254883, "target_z": -1.2109374905111492, "ce": 6.1020402908325195, "genuineness": 4.381826400756836, "identity": 0.5653775930404663, "hypothesis": "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_RANT_reg/249f79c51f131389192892eac1b4916e648ec02670bd28b5773853481f1b53de.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": "24a6c82531db97b5420d933298d5fa0d0bf7c44df564e21bd3fc425b65de9715", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce11f4edc5e8996f7c132ea1e3083392589b2b03bf02393ee61a5bb487c9157c", "generation_seconds_batch": 21.312949602026492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.298274517059326, "target_z": 1.4609375138683203, "ce": 6.125868320465088, "genuineness": 3.7793264389038086, "identity": 0.5436130166053772, "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_RANT_reg/24a6c82531db97b5420d933298d5fa0d0bf7c44df564e21bd3fc425b65de9715.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "2577049b3b6f9f29f1da178f794fa22c5a68b4eb32023f6f14f1d2a9c49be725", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3d03ea3fef3b6d634b4bb509de2ff1ff5ba61622ee229439ae62d90716203f7d", "generation_seconds_batch": 22.14719259995036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9761171340942383, "target_z": 1.9921875868594798, "ce": 5.601652145385742, "genuineness": 3.7040140628814697, "identity": 0.40480494499206543, "hypothesis": "Si trouver la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient !", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2577049b3b6f9f29f1da178f794fa22c5a68b4eb32023f6f14f1d2a9c49be725.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "25eefa40ed333e8763cfa19c8f6a05f06677d6713f3ec21c61423394b56fdbbd", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e2cd63722d81eedeb754ef49ec61840e485cca6707284422dcb00e50be3f22b2", "generation_seconds_batch": 17.377502955961972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9706828594207764, "target_z": -0.36328127518195, "ce": 6.117925643920898, "genuineness": 4.218649387359619, "identity": 0.7069796323776245, "hypothesis": "De deur 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_RANT_reg/25eefa40ed333e8763cfa19c8f6a05f06677d6713f3ec21c61423394b56fdbbd.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": "26164330e259ad80f4d1f5cd7211e70deed4e1eea65b9332933c47f87078cc4b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "059362c7a0cdae430b497a17b2cb7e1644c974eb9619c8e38b57e07d8dfdbaa9", "generation_seconds_batch": 16.696319916984066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7599878311157227, "target_z": 1.0390625328460217, "ce": 5.617537498474121, "genuineness": 3.6789097785949707, "identity": 0.6582412719726562, "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_RANT_reg/26164330e259ad80f4d1f5cd7211e70deed4e1eea65b9332933c47f87078cc4b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "261f732913b89002f262b5ddf881d1ae1e38bf021a0e52d80b4d11ec8ef448a6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "486b6c4b6c348bfbc7b51d4c5e3b89eb1123e2a1fb9c7039e14090bf7e3d07fd", "generation_seconds_batch": 22.15474002296105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.882779836654663, "target_z": 0.3515625153281435, "ce": 6.173524379730225, "genuineness": 3.9801597595214844, "identity": 0.6527047157287598, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/261f732913b89002f262b5ddf881d1ae1e38bf021a0e52d80b4d11ec8ef448a6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "266aef2cba5841a28963f181f3c07f86b9a0accc0f7fa6a487352f11dc8abcf2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdc730cc5a28c302cd99f2e4cad8905c0f05a6f423ba8a1a68bcafc514beb52a", "generation_seconds_batch": 18.233559193089604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0690040588378906, "target_z": 1.2812498627766202, "ce": 6.046441555023193, "genuineness": 3.6067352294921875, "identity": 0.4065896272659302, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/266aef2cba5841a28963f181f3c07f86b9a0accc0f7fa6a487352f11dc8abcf2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "267f1da6a2c980d6975cf5eaefd2323a5245d8817637df7357c6d60402d9eaac", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "962c22ab2d24dde0e0ffd483db510d60be4f716792df0e4e50afe27599c59700", "generation_seconds_batch": 16.84467222506646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.929448366165161, "target_z": 1.1718749576651275, "ce": 5.426913261413574, "genuineness": 3.735394239425659, "identity": 0.7066022753715515, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/267f1da6a2c980d6975cf5eaefd2323a5245d8817637df7357c6d60402d9eaac.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "1d74b228ef9a85ea8cf915b9c6688ef1de3a1f8242f4722dd3e7e8843557a96c", "generation_seconds_batch": 15.710559557075612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24798309803009033, "target_z": -0.9296875343058449, "ce": 6.427689552307129, "genuineness": 4.3127899169921875, "identity": 0.2887783646583557, "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_RANT_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "272abfe9bcfbffc3b6c9f0b6f1094e6c838a874888ef886ab24061fd3f5e0ec6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "3c19d780520f6d1f2f692d79d951b7bae314e7ba4de6e4becc25fc7ae3d9130e", "generation_seconds_batch": 20.973513899021782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906339168548584, "target_z": 1.9375000408750493, "ce": 5.677107334136963, "genuineness": 3.3964879512786865, "identity": 0.39097803831100464, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/272abfe9bcfbffc3b6c9f0b6f1094e6c838a874888ef886ab24061fd3f5e0ec6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "272e845dfef1adec9179ccafb580a8eb4fd0e20e4ca9b310eaf0e33e573b3524", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "370da2878c7554c6a723ff82ab8dc7c54826b8ffeb575feedfa0c4019cfadb06", "generation_seconds_batch": 17.660610841936432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6471638679504395, "target_z": 1.734374870075736, "ce": 5.80021858215332, "genuineness": 3.352555751800537, "identity": 0.5913032293319702, "hypothesis": "Ik heb een kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/272e845dfef1adec9179ccafb580a8eb4fd0e20e4ca9b310eaf0e33e573b3524.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "279a06d05354a86f1df741980e43c6572b2eb5935ff6d4c3bd01b93c448de934", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "010d3ef0db0446624e8f27dfee3956c63ed82e58f77dad0a5f051c304d2182ec", "generation_seconds_batch": 20.461893558036536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.13081121444702148, "target_z": -1.226562503649558, "ce": 5.839932441711426, "genuineness": 4.011539936065674, "identity": 0.4795111417770386, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/279a06d05354a86f1df741980e43c6572b2eb5935ff6d4c3bd01b93c448de934.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "2850ca60d5b327e5a973ea7f18f7c7757f45d0a1879b383b41fca4d2c51bf4bb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "12d10f2cc5c0b39b1a32e9f3b101fc4a6118960b8bba4db4a0e0235f76e51464", "generation_seconds_batch": 21.453761265962385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08849072456359863, "target_z": -1.0546875459844305, "ce": 6.300607204437256, "genuineness": 4.281409740447998, "identity": 0.36838674545288086, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2850ca60d5b327e5a973ea7f18f7c7757f45d0a1879b383b41fca4d2c51bf4bb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "291eab59f82949fda228f890cb94ae754838aefce29073831c7f179fe3761616", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88c60577bf385c40974ee071f4ae969fd786e3a9aeaeb5a7f9cb0a489901acc3", "generation_seconds_batch": 23.661306763999164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3979573249816895, "target_z": 1.5390625795603639, "ce": 6.181467056274414, "genuineness": 3.7416701316833496, "identity": 0.6188651919364929, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/291eab59f82949fda228f890cb94ae754838aefce29073831c7f179fe3761616.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "29cf4b8e7d33f228854828e9b656344deb772627321936d131273368925729c1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "808ce716c7da281b94062c39ba37fb8b7b478dcf590c0049a8742012c76e8bad", "generation_seconds_batch": 24.172602790989913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.500812530517578, "target_z": 0.8359373620467085, "ce": 6.086154937744141, "genuineness": 3.8420867919921875, "identity": 0.6386768817901611, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/29cf4b8e7d33f228854828e9b656344deb772627321936d131273368925729c1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "2a4781f7edfdb5cae477cfb61149c93668839d4acd86676b29033e37f323364a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ff11b1bb2c4a06fd98c91ccd42fdc821baa9b77b931d4739f8761d0e67972cb8", "generation_seconds_batch": 20.58774489501957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.020524263381958, "target_z": -0.32421874233592823, "ce": 6.117925643920898, "genuineness": 3.7918784618377686, "identity": 0.4790223240852356, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2a4781f7edfdb5cae477cfb61149c93668839d4acd86676b29033e37f323364a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "2a9082eeee952a7f0d35ff22e5ddb4f5dec1f1610f83c514b09956db5f9faaee", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6072496a66ab7ae73c1f2eaf31a01ca5d0e8542ab165052029bbcc31fcb3d65e", "generation_seconds_batch": 23.547742606024258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.187492847442627, "target_z": -0.19335935073043947, "ce": 6.165581703186035, "genuineness": 4.055472373962402, "identity": 0.5412246584892273, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2a9082eeee952a7f0d35ff22e5ddb4f5dec1f1610f83c514b09956db5f9faaee.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "2aa0cd187a08a72de2f2290aba8a08b93dc3127a40f1d78fce21516de82ee32d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a5a407d50ad52925a049bc527c4ed42cfa96c1a5103db812167f8850d3629ba9", "generation_seconds_batch": 17.06083371094428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.228496551513672, "target_z": 1.4062499678838898, "ce": 5.9431867599487305, "genuineness": 3.7291181087493896, "identity": 0.38595202565193176, "hypothesis": "J'ai trouvé la 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_RANT_reg/2aa0cd187a08a72de2f2290aba8a08b93dc3127a40f1d78fce21516de82ee32d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "2aae699f9549ecc08e0efd222a00e9caafdc3713c846c19d8589646e34584bc5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "eba7f3fe1ad6ca252705a2a112b13da75892fe7fe64a7fb826f7b582e55ad23a", "generation_seconds_batch": 24.284205751959234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0291311740875244, "target_z": 1.250000023357171, "ce": 5.982900142669678, "genuineness": 3.440420389175415, "identity": 0.5606672167778015, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2aae699f9549ecc08e0efd222a00e9caafdc3713c846c19d8589646e34584bc5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "2ad0b8864c27f49f84b39bc169e18dafc8e59adb69ac1c72c720c728d6660310", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "011e33818320a05823b5c94e3b290cd736286afb0cbc9b287a9657309782223f", "generation_seconds_batch": 19.742142562987283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.447798728942871, "target_z": 1.5781251124063855, "ce": 6.062326908111572, "genuineness": 3.5722172260284424, "identity": 0.5456150770187378, "hypothesis": "J'ai trouvé la 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_RANT_reg/2ad0b8864c27f49f84b39bc169e18dafc8e59adb69ac1c72c720c728d6660310.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": "2b04280742b570fc085feb02fd6d5c78f50171fcc38e2003aba87a8cb16280c5", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f0815c0f9329bf8ef2c1397c0b33f9cc4876caaaaa28dc23652193a34997d53a", "generation_seconds_batch": 21.52681179495994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5506539344787598, "target_z": 0.8749998948927303, "ce": 6.229123115539551, "genuineness": 3.5345609188079834, "identity": 0.39514249563217163, "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": 0.120000000000001, "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_RANT_reg/2b04280742b570fc085feb02fd6d5c78f50171fcc38e2003aba87a8cb16280c5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "2c0baf937cf0749b6ffc03b73e1bb6e7c4b14988dbc7354442ca1219f5e8bf54", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "83d8428ae6f61f3982e6136544c4d5aee264b628894186a2409437db2dc89866", "generation_seconds_batch": 23.354401162010618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.852874994277954, "target_z": 0.32812499562053044, "ce": 5.998785495758057, "genuineness": 4.030367851257324, "identity": 0.6427353620529175, "hypothesis": "La porte 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_RANT_reg/2c0baf937cf0749b6ffc03b73e1bb6e7c4b14988dbc7354442ca1219f5e8bf54.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": "2c2d45a572c299cf461d7918fb205b4b5b3d4f0322705f79f121b7d04c0669ee", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "9c4b544bb6fb5220ca708ae1f4b53b5e6f600646e13e3ea23a7a9fc0a2cb94da", "generation_seconds_batch": 21.677049256977625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.637195587158203, "target_z": 1.7265623635065317, "ce": 6.23706579208374, "genuineness": 3.7542223930358887, "identity": 0.5754101276397705, "hypothesis": "Seine Nachricht kam, als ich schon weg war! Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können!", "wer": 0.041666666666666664, "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_RANT_reg/2c2d45a572c299cf461d7918fb205b4b5b3d4f0322705f79f121b7d04c0669ee.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "2ccc6053cb281a9d425541a4ec4f72811039e7b5bb9c8a8e92a477ac6cefcbad", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "d4fb345c4faf4762c3004db766eac8faa550fc0ad87c8c27b60ee733e558cc19", "generation_seconds_batch": 19.914919547969475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2484331130981445, "target_z": 1.4218749810222986, "ce": 5.303802013397217, "genuineness": 3.8546390533447266, "identity": 0.13879969716072083, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten!", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2ccc6053cb281a9d425541a4ec4f72811039e7b5bb9c8a8e92a477ac6cefcbad.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "2cf3bb4ccb0d5630b0fda81b631ca9d2376da539d991a9f2caf2ea56c04b6e54", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0bcc49f08b436b437d5b7012d21d7b87428ccef11559f01073207a92ae4a1bba", "generation_seconds_batch": 16.026484497939236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.145577430725098, "target_z": 2.1249998248212174, "ce": 5.514282703399658, "genuineness": 3.7291181087493896, "identity": 0.37963926792144775, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2cf3bb4ccb0d5630b0fda81b631ca9d2376da539d991a9f2caf2ea56c04b6e54.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "2d134d3a4e97faa4e088b43eabb1a1dba7229db5b19062c35ff8a07004d6bfc0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d68fcc77c81654b57647a3e26be4d55e0effdcfa09b58e477e32f792b2a8a1c6", "generation_seconds_batch": 20.057710101013072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0429528951644897, "target_z": -0.30664060255521847, "ce": 6.157639026641846, "genuineness": 4.042920112609863, "identity": 0.7167629599571228, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2d134d3a4e97faa4e088b43eabb1a1dba7229db5b19062c35ff8a07004d6bfc0.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": "2d4f64b455fbcc42659bdbb4b4d93cf40a65cb834632111177f2eb6e3bd4e034", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4277c4444ab2c014ce4436b9b53e1e8bf5938bd7858ba23ed8fefbeffb0a4bce", "generation_seconds_batch": 15.53976936894469, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1886234283447266, "target_z": 1.3749999416070724, "ce": 5.871703147888184, "genuineness": 3.660081624984741, "identity": 0.46517911553382874, "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_RANT_reg/2d4f64b455fbcc42659bdbb4b4d93cf40a65cb834632111177f2eb6e3bd4e034.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "2e2945db5c8a1792292b0af238e44e1ac77ff264c99fea6467bb3800134cc617", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c4303eb92ece27d07d0171f11ae187466ffc964db243fdd3d358aee19a5e673a", "generation_seconds_batch": 21.8986858939752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.387989044189453, "target_z": 1.5312500729911596, "ce": 5.6334228515625, "genuineness": 3.4843525886535645, "identity": 0.38872167468070984, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2e2945db5c8a1792292b0af238e44e1ac77ff264c99fea6467bb3800134cc617.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "2e34b9cd17b9322bf7adc0f67c1c038bd96879fb0194f78d0d6e64a5c90ff165", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b74f93489ef1fb8427d251998623eec95781b92d80362a4e27a8c969f9138818", "generation_seconds_batch": 20.5233445399208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7667832374572754, "target_z": 1.8281249489061884, "ce": 5.974957466125488, "genuineness": 3.7291181087493896, "identity": -0.008501827716827393, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2e34b9cd17b9322bf7adc0f67c1c038bd96879fb0194f78d0d6e64a5c90ff165.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": "2e4d7b20c62825f62ef435efd48e8b2403ed2766cc16f921250554e58f39478d", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ed8cfc637e51bc18189adbc6c46c9347f1a9d27ddac86f2939a89e9a7c8a5dd", "generation_seconds_batch": 20.790251467959024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.286494731903076, "target_z": 0.6679686576661832, "ce": 6.078212261199951, "genuineness": 3.6914620399475098, "identity": 0.5246155261993408, "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_RANT_reg/2e4d7b20c62825f62ef435efd48e8b2403ed2766cc16f921250554e58f39478d.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": "2f5197c313b9cf7363c6f1bcab8397bc06e5d085a0e95e6e80e02f0c1b64b0f6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "18044b9fc4835bde256cff276e22303e0f4b8f3496b099c04681b1314ad60b34", "generation_seconds_batch": 17.184968113084324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8864026069641113, "target_z": 1.9218750277366405, "ce": 5.756534099578857, "genuineness": 3.6412534713745117, "identity": 0.2980341911315918, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2f5197c313b9cf7363c6f1bcab8397bc06e5d085a0e95e6e80e02f0c1b64b0f6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "2fa4fd257d16e73f8e49450a9ca0e842112c393a072820f1d0941150f2933907", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6c56aa71584ce27179cbd7ca4c8c19d4d255229d7911c413642d683253a2b94d", "generation_seconds_batch": 16.30183433101047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8560477495193481, "target_z": -0.4531250072991159, "ce": 6.213237762451172, "genuineness": 4.118232727050781, "identity": 0.5681655406951904, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/2fa4fd257d16e73f8e49450a9ca0e842112c393a072820f1d0941150f2933907.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "304387fc4e7799a284be004ef484b3bc92c2cd0aada072a049271b1b7be7c9e7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "9fe18a37f51cfa1f8405f0004c439f53fcf489cf0d3489cff5587f6a4321033f", "generation_seconds_batch": 25.648465163074434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627227306365967, "target_z": 1.7187498569373274, "ce": 5.903473377227783, "genuineness": 3.7416701316833496, "identity": 0.49887800216674805, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/304387fc4e7799a284be004ef484b3bc92c2cd0aada072a049271b1b7be7c9e7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "313249bf3f2f7b1559065275baaa14ca6d8b3872439e43d1c40ec39af3baa3c7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cb527da1377410548120aedcafa98317311e319b9f694fcedd9f35c5ee6bc7c9", "generation_seconds_batch": 24.767288375995122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6535096168518066, "target_z": 0.17187505109381165, "ce": 6.022613525390625, "genuineness": 4.030367851257324, "identity": 0.6096351146697998, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/313249bf3f2f7b1559065275baaa14ca6d8b3872439e43d1c40ec39af3baa3c7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "3152948a4eaedb51f08f76adec65cc1217f293c8edaa31b9297a71bf8205a783", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ee1de34bb699a64f342d2fe9a37b9a94cd0b6c2a0e7718d81e16f5d7be61a532", "generation_seconds_batch": 24.076583515037782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.401129961013794, "target_z": 0.7578124832120333, "ce": 5.8875885009765625, "genuineness": 3.7040140628814697, "identity": 0.5356758832931519, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3152948a4eaedb51f08f76adec65cc1217f293c8edaa31b9297a71bf8205a783.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "3167c428e1f050d081d64b8e7a3c30eff09a0904da27d3f3574a233d3a71dbdc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "247ee9c2acc2e546f4b4b8e5c8459b5d07825b125bfab587ab34d43efb6eb9ae", "generation_seconds_batch": 24.2149029240245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348115921020508, "target_z": 1.5000000467143422, "ce": 5.788304805755615, "genuineness": 3.7793264389038086, "identity": 0.11674656718969345, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3167c428e1f050d081d64b8e7a3c30eff09a0904da27d3f3574a233d3a71dbdc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "317689da91074ec1df04658183acc8e28e2b9af12b94ad0c6f235d23661262fa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "aadef0894cd1bc50ce39eb55c34264923bf4de13defdffeac05e593b48be3603", "generation_seconds_batch": 20.200889728963375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07852256298065186, "target_z": -1.0624999591249507, "ce": 6.141753673553467, "genuineness": 4.287685871124268, "identity": 0.5078796744346619, "hypothesis": "La porte 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_RANT_reg/317689da91074ec1df04658183acc8e28e2b9af12b94ad0c6f235d23661262fa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "317d0f1ea0a5a02b43e9ec2c7361aa0116d5e10349ddc01c703a82d91b880e8e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d128aff9867d91b6c7b1d86eaf58608d9a801f3af8e924f1f3d1f22e825c6f5", "generation_seconds_batch": 21.449753346969374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6467140316963196, "target_z": -0.6171875051093811, "ce": 5.748591423034668, "genuineness": 4.030367851257324, "identity": 0.6397149562835693, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/317d0f1ea0a5a02b43e9ec2c7361aa0116d5e10349ddc01c703a82d91b880e8e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "31f999c6df3cd9792e1424ee9e90804514f362ae88846150cd082fdb4d73a901", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "591bc77a22cac639cc3aa7f3c8506e526c11326bc514c3c5b14943d8516d1819", "generation_seconds_batch": 15.423976648016833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4459872245788574, "target_z": 0.792968762773453, "ce": 5.95112943649292, "genuineness": 3.886019229888916, "identity": 0.716111421585083, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/31f999c6df3cd9792e1424ee9e90804514f362ae88846150cd082fdb4d73a901.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "3225a0418fe0491863d4f078b9ba44a79f0b04f89ef8d90105490e9d8a2c6198", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b91afa0012293b5a166746fd9668534b601ed22548e4c2a55ac0e0e0341b7c8a", "generation_seconds_batch": 21.9715138199972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7681447267532349, "target_z": 0.2617187832109776, "ce": 6.181467056274414, "genuineness": 4.074300289154053, "identity": 0.7166658043861389, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3225a0418fe0491863d4f078b9ba44a79f0b04f89ef8d90105490e9d8a2c6198.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "32424561fc3decd435cf6e19766805d66aedd67623d98ba0ad5514b4b140c76e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4e7316851528849b3c2e9ddf33d5110b99155becff18ceab81ac41d70dc7720e", "generation_seconds_batch": 20.93573178409133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.82342004776001, "target_z": 2.6562498978123767, "ce": 6.117925643920898, "genuineness": 3.355693817138672, "identity": 0.17945101857185364, "hypothesis": "La porte 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_RANT_reg/32424561fc3decd435cf6e19766805d66aedd67623d98ba0ad5514b4b140c76e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "32787fa80a53d64d2f5ff17bbb7baea1b03f2117f0e4ee8adaa7b1b091f26e2b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "974cebddcf6e0b7cc28fa6e79b19af6020a55c37f44940fdd5f574cf400756f3", "generation_seconds_batch": 19.820723601034842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7414126396179199, "target_z": -0.5429687394162819, "ce": 6.149696350097656, "genuineness": 3.9801597595214844, "identity": 0.4661373496055603, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/32787fa80a53d64d2f5ff17bbb7baea1b03f2117f0e4ee8adaa7b1b091f26e2b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "32de633f541291bae7e0af729ee0940c8de152bbe086a3dbee50cdaba599426e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9ae16aea6b9fa44f3b35d545bbc1eeb153ff07c04b32172b90232ac145760d3a", "generation_seconds_batch": 22.31954799592495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5320789217948914, "target_z": -0.707031237226547, "ce": 5.839932441711426, "genuineness": 4.287685871124268, "identity": 0.3953306972980499, "hypothesis": "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_RANT_reg/32de633f541291bae7e0af729ee0940c8de152bbe086a3dbee50cdaba599426e.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": "332a4ff6e69666c1cc2222f07b77bb33723270e5b84900b6f6d43c946fadc03f", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57f0ceee7606e64d42c80f8ed94aaf6850db1da1fc0a9e0dbf7ffed9b4b67be2", "generation_seconds_batch": 23.706111781997606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.77639102935791, "genuineness": 3.5094566345214844, "identity": 0.5302140712738037, "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_RANT_reg/332a4ff6e69666c1cc2222f07b77bb33723270e5b84900b6f6d43c946fadc03f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "337de1c8c71cc755cdc6c2c2f1ca1dbddb541d5f20d84b4d6b0f34da041ea896", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c79fe6b4101549b008c71425c06f959d11186d42f93a4b60345dcabb9a08c1ab", "generation_seconds_batch": 16.29250151803717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.364879608154297, "target_z": 2.296874969343713, "ce": 5.02977991104126, "genuineness": 3.587907314300537, "identity": 0.6893798112869263, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/337de1c8c71cc755cdc6c2c2f1ca1dbddb541d5f20d84b4d6b0f34da041ea896.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "33a17b06504e2687d3e8f2eaf2281d6a2557a984ff391ff44fcf77320f8a0e18", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6f25acb8f95dee591975ae346c2d7b838d0957fb1c8fedff4c841c3599b40418", "generation_seconds_batch": 18.233559193089604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1487503051757812, "target_z": 1.343749915330255, "ce": 5.772419452667236, "genuineness": 3.6098732948303223, "identity": 0.5890242457389832, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/33a17b06504e2687d3e8f2eaf2281d6a2557a984ff391ff44fcf77320f8a0e18.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "33ab500fa750f3bc70564f7cda6bdd120ce06e85a2eb16efbf4d3433b3acf50f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c87728d9a69cb36c7f8c412eadde160fce7bf838e39386438f980d0808915852", "generation_seconds_batch": 16.84063144400716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.072176933288574, "target_z": 0.49999995328565794, "ce": 5.959072113037109, "genuineness": 3.823258876800537, "identity": 0.5988024473190308, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/33ab500fa750f3bc70564f7cda6bdd120ce06e85a2eb16efbf4d3433b3acf50f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "33db977ba5b1badb82f21bbc11a40ccf19a27a8231eb8c129bd5da4e740af2cd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f113881cb47b927b9bbbb9dd0bbce40411eb40472fe594334875834ab703e5d9", "generation_seconds_batch": 21.235691434005275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0155401229858398, "target_z": -0.32812499562053044, "ce": 6.181467056274414, "genuineness": 3.823258876800537, "identity": 0.32658907771110535, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/33db977ba5b1badb82f21bbc11a40ccf19a27a8231eb8c129bd5da4e740af2cd.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "0f7437fc7fd4980d392f8bd5faaae6e59f124f493a112eb048a119375ec73efa", "generation_seconds_batch": 17.741287630982697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5869044065475464, "target_z": -0.6640624978102652, "ce": 6.141753673553467, "genuineness": 4.444586753845215, "identity": 0.6544681787490845, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine Houtenhek. Ik faktens ik wacht aan de andere kant.", "wer": 0.19047619047619047, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "33ffe07407edcd571991ec6dea45e43a654b803d699a2610982c7b3ec3b8b9c0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1b3a4f06646141240042efa3bb85c2f802b617997ee188fe2a65bea1bcb618a", "generation_seconds_batch": 15.703654404962435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6485254764556885, "target_z": 0.16796879780920948, "ce": 6.070269584655762, "genuineness": 4.086852550506592, "identity": 0.7438035011291504, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/33ffe07407edcd571991ec6dea45e43a654b803d699a2610982c7b3ec3b8b9c0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "3400d491d42286674847a6d3114538f83eec8244893b7536b7dcb61b5b3466c8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cd946aee5da4245465fee53d3c390a6e367491710ccd64262db517315beef85c", "generation_seconds_batch": 25.651038215029985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5357017517089844, "target_z": 0.8632813218962921, "ce": 5.645336627960205, "genuineness": 3.6789097785949707, "identity": 0.6238952875137329, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3400d491d42286674847a6d3114538f83eec8244893b7536b7dcb61b5b3466c8.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": "34231906163d6c7a25ab50c71dac89380f3b59bfb0a62705a9c4551fa2d337fa", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd79bfeedc5216139cf9e60a0f319ddaf000bdcb97e504cede105d9a2fe7cd27", "generation_seconds_batch": 25.612769053084776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.417893886566162, "target_z": 1.5546875926987727, "ce": 6.054384231567383, "genuineness": 3.7542223930358887, "identity": 0.2217528522014618, "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_RANT_reg/34231906163d6c7a25ab50c71dac89380f3b59bfb0a62705a9c4551fa2d337fa.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": "345ec1b5edbde5198e0a27e0b473868fd002b2646ca8611a3581a1a6c36c524b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fd758281e4b7976b06e20d9ef9cef119706e5c9aed191e86aaa333aac620de3", "generation_seconds_batch": 25.65482977393549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5474815368652344, "target_z": 1.6562501780984291, "ce": 6.0146708488464355, "genuineness": 3.8169827461242676, "identity": 0.24463066458702087, "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_RANT_reg/345ec1b5edbde5198e0a27e0b473868fd002b2646ca8611a3581a1a6c36c524b.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "c5e61921d4beabe9c29e5d5ccde93075b0ae2463e6e01d8b589d02bf7ea97c8f", "generation_seconds_batch": 15.96823422703892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4274120330810547, "target_z": -0.7890625094888507, "ce": 6.348263263702393, "genuineness": 3.8922953605651855, "identity": 0.3004019558429718, "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_RANT_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "3494a1c7801622be401c7f68ab17d635fe3e0afbc5342be1f1d78a0ac63beeb4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ab67148b94b8c39c5cb45d7d72cdc40c67de02f58f9c8a92ebbbab71145ca0aa", "generation_seconds_batch": 20.486734170001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.163818120956421, "target_z": -0.2119141005466419, "ce": 6.252951145172119, "genuineness": 3.9801597595214844, "identity": 0.40869686007499695, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3494a1c7801622be401c7f68ab17d635fe3e0afbc5342be1f1d78a0ac63beeb4.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": "34eba23f8a9445303072d47fdf16216d307bb829bbd5d74f43e7af20eac506cd", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "2530689bc33594e127b541325dc9698909eb609483d877848217d4ae95e852c1", "generation_seconds_batch": 21.677049256977625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049067735671997, "target_z": 1.2656250364955797, "ce": 5.927301406860352, "genuineness": 3.9048471450805664, "identity": 0.26006951928138733, "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": 0.120000000000001, "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_RANT_reg/34eba23f8a9445303072d47fdf16216d307bb829bbd5d74f43e7af20eac506cd.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "3517377442df5f22ece4e3683fba7e935ecbf38a1257b28ff419730b36b5f028", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "376d46d4304dbb1b055415aa271eb0baec1651e4d08a6e3b11e0e91039b51ad0", "generation_seconds_batch": 25.029188922024332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6471638679504395, "target_z": 1.734374870075736, "ce": 5.855817794799805, "genuineness": 3.393350124359131, "identity": 0.40196430683135986, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3517377442df5f22ece4e3683fba7e935ecbf38a1257b28ff419730b36b5f028.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "3522ad23c53a2b9401b9f1571642387c2b335c65804d2040156d116e75a38adc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "696905df5773d3fbfcf236e7e70115009dcf003fd538822a3552c377237f72de", "generation_seconds_batch": 20.855221818899736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5656065940856934, "target_z": 0.8867188416039051, "ce": 5.311744689941406, "genuineness": 3.672633647918701, "identity": 0.39000940322875977, "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_RANT_reg/3522ad23c53a2b9401b9f1571642387c2b335c65804d2040156d116e75a38adc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "3575f5ea48986f65309e01b6056b486268ea43540048d22df1658c66a1536e5b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "e7d0191556c67f6816683e11a8380f435c54db6b4e1a572bb61461c3939fecd0", "generation_seconds_batch": 25.183253731927834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749863147735596, "target_z": 0.4238281076645996, "ce": 6.0146708488464355, "genuineness": 3.7165660858154297, "identity": 0.3806631863117218, "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_RANT_reg/3575f5ea48986f65309e01b6056b486268ea43540048d22df1658c66a1536e5b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "358d3f8e9024304fcefbbdee0aaea4af7f4b02baed86a27a1e4c1a9a05b21f14", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2c9a5eb8fab9d967e715ffa7f5b3e0a8e5d3ee2d0b8657ddfdfaf63582b87299", "generation_seconds_batch": 24.843240538961254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1625721454620361, "target_z": -0.21289061715345037, "ce": 5.911416053771973, "genuineness": 4.093128204345703, "identity": 0.5905808210372925, "hypothesis": "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_RANT_reg/358d3f8e9024304fcefbbdee0aaea4af7f4b02baed86a27a1e4c1a9a05b21f14.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "35a35622a827a89bc15ef3e17e0937a67dafcf3401a15131ecd336995aeac6e5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "512b1a7fcfec7dd753e54423ab7b800dccfac6ce2ec88a3a2d07842b8e7b2402", "generation_seconds_batch": 25.832968196016736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.407925605773926, "target_z": 1.5468750861295681, "ce": 6.062326908111572, "genuineness": 3.653805732727051, "identity": 0.5392067432403564, "hypothesis": "J'ai trouvé la 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_RANT_reg/35a35622a827a89bc15ef3e17e0937a67dafcf3401a15131ecd336995aeac6e5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "3686ef41d0e60371f4ce30406e9e0469e95b287d9d3141e1e2b9a19037c5166e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cb75c6d633bbaaa23573d9a02ea716b6f8fdb89325203619390408fb012f76b6", "generation_seconds_batch": 22.41117427602876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7581764459609985, "target_z": 0.25390627664177323, "ce": 5.665193557739258, "genuineness": 3.9299514293670654, "identity": 0.6186937093734741, "hypothesis": "La porte 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_RANT_reg/3686ef41d0e60371f4ce30406e9e0469e95b287d9d3141e1e2b9a19037c5166e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "36b655911f31d597b67d48bb9d17a8aabe07430280d72b43602bad6e24aad0ee", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0f36f93b73b081f642c2ee2dba23e0f9670fb3263b34c02ee559e6fd8e4666ca", "generation_seconds_batch": 16.791696242056787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4278621673583984, "target_z": 1.562500099267977, "ce": 5.895530700683594, "genuineness": 3.4623863697052, "identity": 0.6454862356185913, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/36b655911f31d597b67d48bb9d17a8aabe07430280d72b43602bad6e24aad0ee.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "3794801eeb47cc0c0b56b9cbddb6e12c460f5552379f72b6988ccacaf03d8636", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "739939f9d84259dffba4e9bcd991adca798293bc344b40c8116bcf0eba6260ad", "generation_seconds_batch": 21.953229532926343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.211732864379883, "target_z": 0.6093750452545189, "ce": 6.141753673553467, "genuineness": 3.810706615447998, "identity": 0.7135189175605774, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3794801eeb47cc0c0b56b9cbddb6e12c460f5552379f72b6988ccacaf03d8636.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "37bc0765b2b10bc9afb379c441e51a13f1c812ec98a28dc94eecc9271ae1117c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "279158fc4abee9044c0e06770c67ca0b0561c25c9363ca5482db93724a4ca69e", "generation_seconds_batch": 15.08883035206236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6759381294250488, "target_z": 0.1894530974458373, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.5456828474998474, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/37bc0765b2b10bc9afb379c441e51a13f1c812ec98a28dc94eecc9271ae1117c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "37c7219b6f21d53746ec2bb3b338b2767fa64896950250181a33c0336fe587af", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c4dcb3831ff081e688f18c87ad060536601c8fbd71908f4f126e3b425aa32b7", "generation_seconds_batch": 16.78206684801262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.846529483795166, "target_z": 1.8906250014598232, "ce": 5.696964263916016, "genuineness": 3.4812145233154297, "identity": 0.547322154045105, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/37c7219b6f21d53746ec2bb3b338b2767fa64896950250181a33c0336fe587af.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": "381380002e19177a46798e83bad33891bb9bc91b126d600cf7c7460c0977c9a7", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cd65e2a73399669bdedde8c01740bf832198df0f9ae134bb591c73c943018e9", "generation_seconds_batch": 25.42512089491356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3680524826049805, "target_z": 1.5156250598527508, "ce": 6.022613525390625, "genuineness": 3.9425034523010254, "identity": 0.497706800699234, "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_RANT_reg/381380002e19177a46798e83bad33891bb9bc91b126d600cf7c7460c0977c9a7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "385bfbbc7283ada951faf40cddc161ed2800f80215641b6a9865530cf2f32de2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cbd7aa25a8f1f5dce6c470d1039733585ae57479bbe25bf57ad922d2bf4f8384", "generation_seconds_batch": 20.68142329296097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.756814956665039, "target_z": 1.820312442336984, "ce": 5.4666266441345215, "genuineness": 3.5188708305358887, "identity": 0.5418468713760376, "hypothesis": "Ik 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_RANT_reg/385bfbbc7283ada951faf40cddc161ed2800f80215641b6a9865530cf2f32de2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "38d43189525aa49d1ef066786c9b23826d035ea2233a3b443b6cc979d8ec83e5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c0b2d9633f7a491a2c5f4dfe79dc78217c2d7dd68c315d189768cc61e0fed4af", "generation_seconds_batch": 18.538346052984707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.286494731903076, "target_z": 0.6679686576661832, "ce": 5.903473377227783, "genuineness": 3.6412534713745117, "identity": 0.29215705394744873, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/38d43189525aa49d1ef066786c9b23826d035ea2233a3b443b6cc979d8ec83e5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "394222fc06eaad5185d51ad367705395b80cb5f9edafe8a0d82a48aba3d7b68b", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "29cbffd1660c695bb5a13201a6f381978f10d6fcd478221b59dc94e9fc8765ca", "generation_seconds_batch": 18.55227396299597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07852256298065186, "target_z": -1.0624999591249507, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.7085734605789185, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/394222fc06eaad5185d51ad367705395b80cb5f9edafe8a0d82a48aba3d7b68b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "396b2a352d1bbd12b954cc9f554a152aecd4c7c8050c8477cc3367b5b30e8f16", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "dd00fb0df5f3f02e43d77d22ea03f6a33340c3346fa1a904b9553127018f7dab", "generation_seconds_batch": 23.25064266996924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6896445751190186, "target_z": 0.20019534069283534, "ce": 5.748591423034668, "genuineness": 3.9425034523010254, "identity": 0.4728994071483612, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/396b2a352d1bbd12b954cc9f554a152aecd4c7c8050c8477cc3367b5b30e8f16.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "399650c53894f7ba268062e48d6623dd02d646ee2b842633254659c1acc41771", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "da42ce8302b625806898fc1dce0578961f1f9b331c73dfda4b83f4439577b695", "generation_seconds_batch": 23.486754903919064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7799243927001953, "target_z": 1.0546875459844305, "ce": 5.8875885009765625, "genuineness": 3.7667744159698486, "identity": 0.28659766912460327, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/399650c53894f7ba268062e48d6623dd02d646ee2b842633254659c1acc41771.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "39cb8bfe7f616b903f504d4f8c3b7f4dd5c32205e972919e72e5cfb7c34ddf64", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "9a59929ebe76fee70dc3ec1ecef93d757e51acef408b0f7e17426293429fbc33", "generation_seconds_batch": 21.443942278972827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8653353452682495, "target_z": 0.33789062883203586, "ce": 6.038498878479004, "genuineness": 4.105680465698242, "identity": 0.4272908568382263, "hypothesis": "La porte 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_RANT_reg/39cb8bfe7f616b903f504d4f8c3b7f4dd5c32205e972919e72e5cfb7c34ddf64.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "3a1089db4b1cdb0882a3fc1b6fc070686bb70e8a2857cdfb6298984aabfb071d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "070d808132a5ef434dfbf50b2715cdaa0eb88239f6e67c7f5aeac15f1413528a", "generation_seconds_batch": 26.71745335299056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3014473915100098, "target_z": 0.6796876043773581, "ce": 5.657250881195068, "genuineness": 3.7793264389038086, "identity": 0.6495504379272461, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3a1089db4b1cdb0882a3fc1b6fc070686bb70e8a2857cdfb6298984aabfb071d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "3a4fdc655551f05846506ba0f4b1694e1c1537f74aaf5576ce3b40a3f70933dc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d21ac6ef4ea864b2cd42a39df062248d1d65f30dedf2712d5867fa73ff05f56d", "generation_seconds_batch": 15.942274330998771, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.846529483795166, "target_z": 1.8906250014598232, "ce": 5.534139156341553, "genuineness": 3.365107774734497, "identity": 0.38370537757873535, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3a4fdc655551f05846506ba0f4b1694e1c1537f74aaf5576ce3b40a3f70933dc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "3a88b34609d70e34359872c22917521a2341d8ea6eb965f75138b87bb77aacc3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "682dab0584aeaa0a7f469d060da0fca6bcb53a7abc1051ba44d1802bee0edf1f", "generation_seconds_batch": 18.307726435014047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5456700325012207, "target_z": 0.8710938284654964, "ce": 5.33954381942749, "genuineness": 3.6977379322052, "identity": 0.6126495003700256, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3a88b34609d70e34359872c22917521a2341d8ea6eb965f75138b87bb77aacc3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "3aa66781d2a4f9242126e4d9b61b9afea80b605cea5de256a188c6de52a965be", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b17b68c57b7b39d64b3cd3f26703b84ee2706d415d748ad4c5f106f5b3cda11b", "generation_seconds_batch": 23.050366302020848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3680524826049805, "target_z": 1.5156250598527508, "ce": 5.808161735534668, "genuineness": 3.4215922355651855, "identity": 0.45376646518707275, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3aa66781d2a4f9242126e4d9b61b9afea80b605cea5de256a188c6de52a965be.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": "3b301dc7c5984f807e8c8a2dabcf6d7d46f1693381159021bdc67756dd39f889", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b93fc5006ae14b91aab57c45cee8426575d0760040c239f6d6077dee3120da74", "generation_seconds_batch": 23.024942449992523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.507608413696289, "target_z": 1.6250001518216117, "ce": 6.23706579208374, "genuineness": 3.823258876800537, "identity": 0.5822354555130005, "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_RANT_reg/3b301dc7c5984f807e8c8a2dabcf6d7d46f1693381159021bdc67756dd39f889.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "3bce38d5d16f8893caec293762a46b0e62510f088ec5d9f114d94230912f5ac8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d8c0ca18b42d14b9a82771a63889fa19556ac18e56b8837de3dc85c08c8f327b", "generation_seconds_batch": 23.538713185000233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1077466011047363, "target_z": -0.2558594032840743, "ce": 6.229123115539551, "genuineness": 4.319066047668457, "identity": 0.6513561010360718, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3bce38d5d16f8893caec293762a46b0e62510f088ec5d9f114d94230912f5ac8.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "a6ed9a43f2ee647b18a7292f2d11c0e39973c3d3ab7c6a4f141eaac20f6dc6fe", "generation_seconds_batch": 18.24091650999617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47725343704223633, "target_z": -0.7499999766428289, "ce": 6.0146708488464355, "genuineness": 4.369274139404297, "identity": 0.6975505352020264, "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_RANT_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "3cae1b5ec4f7739bbf1e3eb9db302af4eb1dd466cb42ba42b3c1532d8908a197", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "385139067e23ef530304abf509ff9a40a17353f8c9e8fd0c141229bd277d04ee", "generation_seconds_batch": 21.02991037000902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0012236833572387695, "target_z": -1.1250000116785854, "ce": 6.292664527893066, "genuineness": 4.362998008728027, "identity": 0.5923133492469788, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3cae1b5ec4f7739bbf1e3eb9db302af4eb1dd466cb42ba42b3c1532d8908a197.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "3d45331a010b24ea6106b348778f1b8d68c46a4abdf9aada4e5c052862df8846", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "328c28b594e2622ad6ec577866f93c6a4d3049d5f7d4f8c502296dc9addffc6c", "generation_seconds_batch": 24.230891890008934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9400973320007324, "target_z": 0.39648433467238436, "ce": 5.863759994506836, "genuineness": 3.6789097785949707, "identity": 0.4459978938102722, "hypothesis": "Ik 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_RANT_reg/3d45331a010b24ea6106b348778f1b8d68c46a4abdf9aada4e5c052862df8846.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": "3d502a8f31308a101b3ee3e561849e23496ae93a7a71b606c4191635aed531c0", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "258d50e0503379a158ee5b54273768aea34cb4689e471001a14dec4cbb3ba9e4", "generation_seconds_batch": 15.654827943071723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769956111907959, "target_z": 1.0468750394152262, "ce": 5.585766792297363, "genuineness": 3.8546390533447266, "identity": 0.7056678533554077, "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_RANT_reg/3d502a8f31308a101b3ee3e561849e23496ae93a7a71b606c4191635aed531c0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "3ef348461ebbc0895ecc4b6e5effd75b048a30f9612b6732a1564d73ce257729", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "99435630cca4ab79b0845de85c4ce01e3b331f09dbf69ac8770a3b9369ed3513", "generation_seconds_batch": 19.798688989016227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.46456241607666, "target_z": 2.3750000350357565, "ce": 5.5817952156066895, "genuineness": 3.2458629608154297, "identity": 0.6568446159362793, "hypothesis": "De deur staat nog open! Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3ef348461ebbc0895ecc4b6e5effd75b048a30f9612b6732a1564d73ce257729.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "3f7076aed0619e69d99df6ecef7e51364ab8416514e7aa861be56207aaad678c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4dd9b1e9142112a2ef84eede4e216e1b21f0228ff82662ff19370e68e77a3e66", "generation_seconds_batch": 26.521203038981184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.495886206626892, "target_z": 0.04833981432543883, "ce": 6.054384231567383, "genuineness": 3.7416701316833496, "identity": 0.40815469622612, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3f7076aed0619e69d99df6ecef7e51364ab8416514e7aa861be56207aaad678c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "3f96002cc7ed762a9e71f78fc0edce96ef38799411ef1894304d2e415279b85c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2babc850912d9344af7f4beb7f0066a27f70137540d8df196ba55e670af4405e", "generation_seconds_batch": 15.133480048039928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6566823124885559, "target_z": -0.6093749985401768, "ce": 6.260893821716309, "genuineness": 4.2249250411987305, "identity": 0.4029998183250427, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3f96002cc7ed762a9e71f78fc0edce96ef38799411ef1894304d2e415279b85c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "3fa51ef0d6d5e559b76105d4bc8586328231773994c2dd4ac9936948610b8b93", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "070ff57539fb91e2a1ad0c6ae7a9e5c2b2f1074624fdee2e4aa3aee587c66cce", "generation_seconds_batch": 20.118944074958563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7015395164489746, "target_z": -0.5742187656930993, "ce": 6.316492557525635, "genuineness": 4.049196243286133, "identity": 0.5783583521842957, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3fa51ef0d6d5e559b76105d4bc8586328231773994c2dd4ac9936948610b8b93.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "3fcfe2c0e6be2d9c4e3b268691d65ac578002c12bf4472eb5096ee00fe54aa10", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1ab97d07091f97a96aefc0c42fb3744de07cd0e535366e6da964885f0c69c658", "generation_seconds_batch": 22.60050759103615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7282716035842896, "target_z": 0.23046875693416016, "ce": 5.919358730316162, "genuineness": 3.9927117824554443, "identity": 0.5222073793411255, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3fcfe2c0e6be2d9c4e3b268691d65ac578002c12bf4472eb5096ee00fe54aa10.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": "3fdd49d023b17a612226836ac9314c94126cc1f07d5050464a021f958a4005ca", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74648882da8280d37ea3ac9f700966b6c7675d89300383c7299f4dcd570164a6", "generation_seconds_batch": 16.07907515298575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9761171340942383, "target_z": 1.9921875868594798, "ce": 5.6334228515625, "genuineness": 3.653805732727051, "identity": 0.6385800838470459, "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_RANT_reg/3fdd49d023b17a612226836ac9314c94126cc1f07d5050464a021f958a4005ca.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "3ff88c95db92bb743dad1c28ca1e7451654d1eaf5f284171c5527ee2c298657b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ebacc37f71de9a6e0678d46f2a869fa40e104aa57825c24dfd4143d6e230b2a4", "generation_seconds_batch": 21.406599993933924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.185450553894043, "target_z": 2.1562498510980346, "ce": 5.5817952156066895, "genuineness": 3.653805732727051, "identity": 0.1979888379573822, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/3ff88c95db92bb743dad1c28ca1e7451654d1eaf5f284171c5527ee2c298657b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "4085a0d7892d9b0267b96e9489bc8364e0179aff5f807ce149cc9fa64d44a414", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e85fed68fb63b4a40bd62c9f6625ca6da062e4620e1e92774e3d217a51da680d", "generation_seconds_batch": 22.581655811984092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0977783203125, "target_z": -0.26367190985327865, "ce": 6.117925643920898, "genuineness": 3.5816311836242676, "identity": 0.45958274602890015, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4085a0d7892d9b0267b96e9489bc8364e0179aff5f807ce149cc9fa64d44a414.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "40b1b07dff064ad2fa90a5f546e4237c77d2fb143b473acbe8b127e25e68addc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c5ea3dab2d31a82f70c958eb5923a55c354aa3c68eda2bf874af1ab1d1bd16f7", "generation_seconds_batch": 19.692703776061535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5044355392456055, "target_z": 2.406250061312574, "ce": 5.689021587371826, "genuineness": 3.4592485427856445, "identity": 0.6164537668228149, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/40b1b07dff064ad2fa90a5f546e4237c77d2fb143b473acbe8b127e25e68addc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "410351063dceebca565fc0c5ccc48744d1e782302609e6e0a2357e2c30d34fff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a8c8e778ce746d4e98c07137e473ea310d7c5f11f9d053a87aad40a92d0c3fca", "generation_seconds_batch": 21.827090765931644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3680524826049805, "target_z": 1.5156250598527508, "ce": 6.022613525390625, "genuineness": 3.9048471450805664, "identity": 0.5545358657836914, "hypothesis": "La porte 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_RANT_reg/410351063dceebca565fc0c5ccc48744d1e782302609e6e0a2357e2c30d34fff.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "411721c1eaaafb897cc724391b0553c1331c6cb7125f75e6901795127812f31a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9ff62068279773107b1060fec23d6f6f0ba3687e8151203a6b3b49b7db82e543", "generation_seconds_batch": 19.384655141970143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.108877182006836, "target_z": 1.3124998890534376, "ce": 6.133810997009277, "genuineness": 3.6914620399475098, "identity": 0.34945952892303467, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/411721c1eaaafb897cc724391b0553c1331c6cb7125f75e6901795127812f31a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "415e8c3babb5bd7404532be592f2e60982eb392001ba252eb1a4a2f145a43626", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "46f030231bd0847b7fa39968ce6ebbb19890033a241b3be650881044b40cc57f", "generation_seconds_batch": 25.63541465695016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.590527057647705, "target_z": 0.9062499211695477, "ce": 5.998785495758057, "genuineness": 3.7416701316833496, "identity": 0.6927297115325928, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. We've regarded ensemble what it contains.", "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_RANT_reg/415e8c3babb5bd7404532be592f2e60982eb392001ba252eb1a4a2f145a43626.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "4174cb2362919b5ed65d657e88b5c6c07b8920574a29233c90077c4a3f161917", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d62fe8e6d679232733aab3283ba9624841b8abf8f9ae3fe534cd0235346a21c4", "generation_seconds_batch": 25.63541465695016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.387989044189453, "target_z": 1.5312500729911596, "ce": 6.141753673553467, "genuineness": 3.7040140628814697, "identity": 0.6414957046508789, "hypothesis": "J'ai trouvé la petite boîte! Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4174cb2362919b5ed65d657e88b5c6c07b8920574a29233c90077c4a3f161917.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "4209060e4829cd70542055a372344a6d4351e4d1311190e0b3cfd5e5e1d74806", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "dc59cd050b9b4f0c28e4ea652c5ee84371036bf2a226378a44d5a26fa4ac71c3", "generation_seconds_batch": 23.24568514805287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5955114364624023, "target_z": 0.9101563613115182, "ce": 5.967014789581299, "genuineness": 3.9676074981689453, "identity": 0.4302208423614502, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4209060e4829cd70542055a372344a6d4351e4d1311190e0b3cfd5e5e1d74806.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "42f836941c6cc5c0bd4ec54d287669609290aacf21015a38e2970735f2f76543", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3b0b22705bd492650d7e70f4a7398c878fbfc0f7a31b9d2c9024e3ff8c1a0a4a", "generation_seconds_batch": 20.200889728963375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.46230101585388184, "target_z": -0.7617187364966355, "ce": 6.09409761428833, "genuineness": 4.3127899169921875, "identity": 0.7688907384872437, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/42f836941c6cc5c0bd4ec54d287669609290aacf21015a38e2970735f2f76543.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "430df08983d9b386c4ac9238a06c168d8c8866793a5af8a1a6566e2a077566ae", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4d5ab0a1df5d5c8485492980f220c937c56fd92dd1f669289c40e5c45c4354ea", "generation_seconds_batch": 20.72816248901654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1775245666503906, "target_z": -0.20117185729964382, "ce": 6.133810997009277, "genuineness": 4.18099308013916, "identity": 0.7384124398231506, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/430df08983d9b386c4ac9238a06c168d8c8866793a5af8a1a6566e2a077566ae.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "432ef76529bc36791f61888b0b2b3f4f6812ee2dcc5b26a523e89a34494b3e6a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "06b395b5661feeae40c08b26b14a6cd79835f7a453bece9cd5a234c347ade693", "generation_seconds_batch": 24.045199324958958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4148939847946167, "target_z": -0.015136754835004476, "ce": 6.157639026641846, "genuineness": 3.8169827461242676, "identity": 0.3983421325683594, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/432ef76529bc36791f61888b0b2b3f4f6812ee2dcc5b26a523e89a34494b3e6a.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "00527723f1f18f3ba60a0b4bc4e7436e5c4f2aa1de8535b57f61679a0d532be5", "generation_seconds_batch": 17.741287630982697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24299895763397217, "target_z": -0.9335937875904471, "ce": 5.998785495758057, "genuineness": 4.287685871124268, "identity": 0.5931320786476135, "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_RANT_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "438ad26db64d450b8554f17455ad8bb0c368bc15967e0c7a615a414af140baac", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bd23bb3d2724e12dda491705391c2b045e2dab54cd7fc0925ae8bf855ccf953d", "generation_seconds_batch": 21.96912500599865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9262757301330566, "target_z": 1.953125054013458, "ce": 5.418970584869385, "genuineness": 3.352555751800537, "identity": 0.5762487649917603, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/438ad26db64d450b8554f17455ad8bb0c368bc15967e0c7a615a414af140baac.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "43b530dda341238bc43a45ebb6fe4e299d9dacc4f4061898050f08bbbe076951", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "06305ba62cd54449eed7e4c800dfc72122fc7a0670a8df002c17d1d43b189588", "generation_seconds_batch": 20.809345219051465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3507232666015625, "target_z": -0.0654296958027444, "ce": 6.1894097328186035, "genuineness": 4.005263805389404, "identity": 0.6171016693115234, "hypothesis": "Ik 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_RANT_reg/43b530dda341238bc43a45ebb6fe4e299d9dacc4f4061898050f08bbbe076951.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "31a6880554a80a3bc07795f4bc6550575bd5b7a7f45ed1021cbbbadb475b44f2", "generation_seconds_batch": 16.60622680198867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07852256298065186, "target_z": -1.0624999591249507, "ce": 6.09409761428833, "genuineness": 4.375550270080566, "identity": 0.6573020219802856, "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_RANT_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.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": "43f712aed71853fc9869e4e5fae1debbff507aad1a554c184b22c30837ba6945", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cf58f2449a7989d75810df9e2a3cd7d13b55b587e2026e5a71176c7c182a14f", "generation_seconds_batch": 21.21213507698849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9693214893341064, "target_z": 1.203124983941945, "ce": 5.550024509429932, "genuineness": 3.798154592514038, "identity": 0.6710419654846191, "hypothesis": "Elguin 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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/43f712aed71853fc9869e4e5fae1debbff507aad1a554c184b22c30837ba6945.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": "44136fc3ab0fed9fb2dcd447f313e5814ac97fd215e851b52c0ce973201cedcf", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26f0f45ba7237320d4e9495d9d72354916bd24bdd9f9da6abf1844d36d9e1ef4", "generation_seconds_batch": 16.880304710008204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.387989044189453, "target_z": 1.5312500729911596, "ce": 5.756534099578857, "genuineness": 3.565941095352173, "identity": 0.41162335872650146, "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_RANT_reg/44136fc3ab0fed9fb2dcd447f313e5814ac97fd215e851b52c0ce973201cedcf.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": "443daabdcca641b054718f63b07bc9462237267a2d9aa82c913f6ede02108b8e", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c591c53df8c3dcb2bafbe9b0201a402d26928bd0297675c45765b96cc831ad3b", "generation_seconds_batch": 21.52681179495994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.108877182006836, "target_z": 1.3124998890534376, "ce": 6.109982967376709, "genuineness": 3.4843525886535645, "identity": 0.41808879375457764, "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_RANT_reg/443daabdcca641b054718f63b07bc9462237267a2d9aa82c913f6ede02108b8e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "4447bcddae2882833ae0d33bb8b6e738242d49f854439662e201511dea896dca", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "468b18acdbab8a5e96e48cad40b6a44af53d4f475ba7e8791c2d5afe89334788", "generation_seconds_batch": 26.150586207048036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5058544874191284, "target_z": 0.056152320894643185, "ce": 6.165581703186035, "genuineness": 4.1998209953308105, "identity": 0.5621529817581177, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4447bcddae2882833ae0d33bb8b6e738242d49f854439662e201511dea896dca.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "445372f99e9eb126ee9ba1f1cd098ba95978ca3eefaf5910ce84161ca75e8d84", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f27a7f1d0be3dc91402b4bcaf8c467912087af4a77e213abcc70374be9da3679", "generation_seconds_batch": 21.991229766979814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3014473915100098, "target_z": 0.6796876043773581, "ce": 6.09409761428833, "genuineness": 3.8044304847717285, "identity": 0.653765857219696, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/445372f99e9eb126ee9ba1f1cd098ba95978ca3eefaf5910ce84161ca75e8d84.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "447a01afc389dd8b746bec90e86ef7b812475bf1bb1cd3f4bf7d2aeef8ca9cab", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0b28d385a890d5a22795f957acd4a2eca7b934476483686a4b330405f5076c0", "generation_seconds_batch": 23.486754903919064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8995437622070312, "target_z": 1.1484376248148827, "ce": 5.597680568695068, "genuineness": 3.8797430992126465, "identity": 0.3497238755226135, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stohl 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_RANT_reg/447a01afc389dd8b746bec90e86ef7b812475bf1bb1cd3f4bf7d2aeef8ca9cab.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": "4509add9d08c753812de0e31cac75d2d9e57764fbfb2716849fa5e00529eb891", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b230dfe7287db718904a14853a733c8b8a01594c0d963994068a552eb71ed85", "generation_seconds_batch": 20.861759269959293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.045895099639893, "target_z": 2.04687513284391, "ce": 5.5659098625183105, "genuineness": 3.415316104888916, "identity": 0.5957775115966797, "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_RANT_reg/4509add9d08c753812de0e31cac75d2d9e57764fbfb2716849fa5e00529eb891.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "455fb0fb1af1c5e3d35207decd3ef8d36ff1a95c6869335273c6bf0c4e4d301f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "afb91ac41390526d2c09a7fc2d9f3f18823ab532b198f64a9685c1d97c923208", "generation_seconds_batch": 21.82428580301348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0012236833572387695, "target_z": -1.1250000116785854, "ce": 6.006728172302246, "genuineness": 4.432034492492676, "identity": 0.5505142211914062, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/455fb0fb1af1c5e3d35207decd3ef8d36ff1a95c6869335273c6bf0c4e4d301f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "45a4bdcec8925da9fe5ed9ee74f943963a866ca063229ab71273dc805415c342", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5b2ed28c9106f7d7cd9c706d994947a42b866eb0812c3d759caf5d9bc797bcf9", "generation_seconds_batch": 24.073972420068458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9561805725097656, "target_z": 1.976562573721071, "ce": 5.8875885009765625, "genuineness": 3.7793264389038086, "identity": 0.3184424042701721, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/45a4bdcec8925da9fe5ed9ee74f943963a866ca063229ab71273dc805415c342.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "45e4b3940c47b93ab6ebe8736e625236ae0a79ab6c144f90c49f8d2875ecb82b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2d128ebe499356023a938952dd45daaf7fcce2af4ac7cc7f60bcfbc180ce7aca", "generation_seconds_batch": 22.32018464396242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6267774701118469, "target_z": -0.6328125182477898, "ce": 5.982900142669678, "genuineness": 3.948779582977295, "identity": 0.7505117654800415, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/45e4b3940c47b93ab6ebe8736e625236ae0a79ab6c144f90c49f8d2875ecb82b.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "f8e66032fe8132f4c3e48fe3d7b6a8162167b555859c8cfaa377360381a7fea6", "generation_seconds_batch": 15.127355437958613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47725343704223633, "target_z": -0.7499999766428289, "ce": 6.260893821716309, "genuineness": 4.068024158477783, "identity": 0.43205004930496216, "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_RANT_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "4613f8bdcee55fe0e1efb2eb68431ffed6a6f5d63426d6519a82af04be6aae83", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "67dfed35bdb3e649879e6b2f98198f5eaffd89906a3092a7a5c39b931426b8c8", "generation_seconds_batch": 21.5575936390087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0690040588378906, "target_z": 1.2812498627766202, "ce": 5.446769714355469, "genuineness": 3.6977379322052, "identity": 0.5083918571472168, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4613f8bdcee55fe0e1efb2eb68431ffed6a6f5d63426d6519a82af04be6aae83.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "464df0bf241fd1f02d1f5825afc21b34d1e9d9a890835364f7b43e36d4e8a82a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "79d7a1506d9726ae687a34c06fcf0156c91238ee01e3dbacca2d31c2b21ea6dd", "generation_seconds_batch": 24.234159923973493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460939645767212, "target_z": 0.8046875226272594, "ce": 5.95112943649292, "genuineness": 3.8797430992126465, "identity": 0.4851418733596802, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/464df0bf241fd1f02d1f5825afc21b34d1e9d9a890835364f7b43e36d4e8a82a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "47275f9561cbf51553b76378d40931f6909a3a134ec7766c141bf1c3b63ae19b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6a1aa2e6d6ae9898e124d893488e3867076313e44178e50cb1dc597e10a4023b", "generation_seconds_batch": 24.234159923973493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769956111907959, "target_z": 1.0468750394152262, "ce": 6.1020402908325195, "genuineness": 3.578493118286133, "identity": 0.6086610555648804, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/47275f9561cbf51553b76378d40931f6909a3a134ec7766c141bf1c3b63ae19b.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "890ef7910c320d86b5e500f11a2c230da2d7d5cf23d40d8a3ae2e834c0caf9ed", "generation_seconds_batch": 14.769946539076045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.1407794952392578, "target_z": -1.2343750102187623, "ce": 6.109982967376709, "genuineness": 4.2374773025512695, "identity": 0.7035934925079346, "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_RANT_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "47d3aade87df528caac4316d346d64df678b5ff9e877fad7c81181b718f8e4c6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bcb3dfd9247bbf0896f24a74daa9f5bac10275b4959a961b00708386752d3beb", "generation_seconds_batch": 23.0880694600055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0091946125030518, "target_z": 1.2343750102187623, "ce": 5.7128496170043945, "genuineness": 3.478076457977295, "identity": 0.49699556827545166, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/47d3aade87df528caac4316d346d64df678b5ff9e877fad7c81181b718f8e4c6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "480d51bf212fc718d0a8ef150dac1744978ee133a403434ca850e95c3bda8cc9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7dfbced91e8fd5cfc3de79dbbb989dc013602bb00c6a20be4ebd7e3b35c1c5cc", "generation_seconds_batch": 23.050366302020848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5574498176574707, "target_z": 1.6640626846676336, "ce": 5.764476776123047, "genuineness": 3.365107774734497, "identity": 0.3118090033531189, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/480d51bf212fc718d0a8ef150dac1744978ee133a403434ca850e95c3bda8cc9.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a9c33488a98c84bb67a25519229594514e04ed1847d2cefd8d5cc5056e8147c6", "generation_seconds_batch": 17.854703341959976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.09093809127807617, "target_z": -1.1953124773727406, "ce": 6.133810997009277, "genuineness": 4.218649387359619, "identity": 0.6907628774642944, "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_RANT_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "4a5b48b5834b415a826f8ae2b284ab8ff6cfbfdc62f92c6108ffb0b4a7712b05", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6174a30f66db08dcb033c4d69e767f87cde7bbab03bccbdd3cd30f54c1d194d1", "generation_seconds_batch": 16.560774549026974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2783379554748535, "target_z": 1.4453125007299117, "ce": 5.879645347595215, "genuineness": 3.860914945602417, "identity": 0.7463064193725586, "hypothesis": "La porte 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_RANT_reg/4a5b48b5834b415a826f8ae2b284ab8ff6cfbfdc62f92c6108ffb0b4a7712b05.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "4ae61a008964de99feabcf0e587d46cd1f8a42b7d83f498bc56aaaa9e084d034", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "9855035e2e81cb8fa173c32a19c9b91ed62c1661f48fa22167fd40de9696e123", "generation_seconds_batch": 22.183919743052684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6796762943267822, "target_z": 0.192382834123631, "ce": 5.879645347595215, "genuineness": 3.911123275756836, "identity": 0.4896066188812256, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4ae61a008964de99feabcf0e587d46cd1f8a42b7d83f498bc56aaaa9e084d034.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": "4ae660289246d693fa1546ffc969039db9ece9070bf28352bb50728129db40f8", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d0c930078b1092f71285cae093412fb49a95a8799fbe2b7826bd8c8278ecd41", "generation_seconds_batch": 30.375837830011733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.108877182006836, "target_z": 1.3124998890534376, "ce": 5.784333229064941, "genuineness": 3.7667744159698486, "identity": 0.6601535081863403, "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_RANT_reg/4ae660289246d693fa1546ffc969039db9ece9070bf28352bb50728129db40f8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "4b08c1b5c77f70cae32684bf49c1febe8d927ba43f604c77652b83c7537947ef", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d74aa62fabb606d6757fc8f665b78a980be7993e8cd04bb4c4eeebb3cc60d3be", "generation_seconds_batch": 25.59376314491965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4577670097351074, "target_z": 1.58593761897559, "ce": 5.855817794799805, "genuineness": 3.7667744159698486, "identity": 0.3064906597137451, "hypothesis": "J'ai trouvé la 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_RANT_reg/4b08c1b5c77f70cae32684bf49c1febe8d927ba43f604c77652b83c7537947ef.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "4b281a128b3f5abd659a824ab642f3167be94d2d12b5513556eefbfeb0ab6fe4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8cc3704682ee05ffa1f12abac75b585599e1e705936fa14f5d9adb773d5901b3", "generation_seconds_batch": 23.552660661982372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4274120330810547, "target_z": -0.7890625094888507, "ce": 6.117925643920898, "genuineness": 4.036643981933594, "identity": 0.5373265147209167, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4b281a128b3f5abd659a824ab642f3167be94d2d12b5513556eefbfeb0ab6fe4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "4b42b374d0b39467332ee97cda52e36f44d4ecc6f2171a96cbf2cc9b7cdb5af1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "10976a1826b01ea8618ade003b4914d97e5554f2814767a533bcef9a518ce091", "generation_seconds_batch": 19.384655141970143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.371225118637085, "target_z": 0.7343749635044202, "ce": 6.1020402908325195, "genuineness": 3.7918784618377686, "identity": 0.2936568260192871, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4b42b374d0b39467332ee97cda52e36f44d4ecc6f2171a96cbf2cc9b7cdb5af1.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": "4bb1633378ecc7bd50e6bcfeecdcaebdefa3b45713cea19bb9f1be80ef44404f", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f28faf608f75194e542e18fdcc8cc31caa26e3b82b3bbb8036c4634fab127d3", "generation_seconds_batch": 16.880304710008204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5275449752807617, "target_z": 1.6406251649600205, "ce": 5.871703147888184, "genuineness": 3.4937667846679688, "identity": 0.33503350615501404, "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_RANT_reg/4bb1633378ecc7bd50e6bcfeecdcaebdefa3b45713cea19bb9f1be80ef44404f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "4c04240bb0066506197832c4f23f0dcced105c116f7b1ee752b9aa3d9fd6c7cc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "350e2c53156c5586ae9ae47e30e513226f91dea9f33f14b661341bbc72a90ee7", "generation_seconds_batch": 19.798688989016227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.02595853805542, "target_z": 2.0312501197055015, "ce": 5.601652145385742, "genuineness": 3.3870739936828613, "identity": 0.6187965869903564, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4c04240bb0066506197832c4f23f0dcced105c116f7b1ee752b9aa3d9fd6c7cc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "4c199ec2cfd7ff38a66035c09f95434f1e27e72811b3cf869d4ef4cd6e5adc2b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "27cdb10d07cc627c6b80509926981afd4a548a48e66c99e60efa97ee95308712", "generation_seconds_batch": 22.434583211899735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.387989044189453, "target_z": 1.5312500729911596, "ce": 5.462655067443848, "genuineness": 3.653805732727051, "identity": 0.2504984438419342, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4c199ec2cfd7ff38a66035c09f95434f1e27e72811b3cf869d4ef4cd6e5adc2b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "4cdc12f026d6d52777ef73057d6fa808b3f620e16959918c99b7173c1715fcb8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97176036e491f77b91905ede473667f778cb581e1575d69b96d3de9eceb79189", "generation_seconds_batch": 20.5233445399208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.305069923400879, "target_z": 2.249999929928487, "ce": 6.0305562019348145, "genuineness": 3.5565271377563477, "identity": 0.5331803560256958, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4cdc12f026d6d52777ef73057d6fa808b3f620e16959918c99b7173c1715fcb8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "4d96b4c7f8f28767babcaea14f2a8f2274d00f8969dd8439fb516c0c632a2866", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a6c5a233faaa6126387aa37d00ba3f9f3ce2cf1611923d8b32ec2562add11636", "generation_seconds_batch": 18.01342276192736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19814181327819824, "target_z": -0.9687499737231826, "ce": 6.292664527893066, "genuineness": 4.369274139404297, "identity": 0.7762717008590698, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4d96b4c7f8f28767babcaea14f2a8f2274d00f8969dd8439fb516c0c632a2866.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": "4e601972555868ae23548054953c8d0966cf79005d541d737d7c90ea705b02d8", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac858a621621a596a9107220ed65496ccaa872df1df8d71476851c6b62288fe8", "generation_seconds_batch": 16.842396187013946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.02595853805542, "target_z": 2.0312501197055015, "ce": 5.649308204650879, "genuineness": 3.515732765197754, "identity": 0.6269245147705078, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la daarna naast het raam. Neem ze mee als je weggaat!", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4e601972555868ae23548054953c8d0966cf79005d541d737d7c90ea705b02d8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "4e782da3d763ab13f4ae4c148aaba6c23a349b728f2fa1a468149aa454dcbdd9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "91988648b504bf8745aa3241af21059f87ada8bcc636857b4b4b13bd051cf4d7", "generation_seconds_batch": 20.63676712499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.507608413696289, "target_z": 1.6250001518216117, "ce": 5.808161735534668, "genuineness": 3.5094566345214844, "identity": 0.7355074882507324, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4e782da3d763ab13f4ae4c148aaba6c23a349b728f2fa1a468149aa454dcbdd9.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": "4eb46370c3a2206135f4e61d94761250ef48b135b824ab4c2d58a900b2b87d15", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1483790cc24618a83cce18db412baa4480ef3a30ee742506bdd98325cf650bc7", "generation_seconds_batch": 15.69900585594587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 5.685050010681152, "genuineness": 3.8169827461242676, "identity": 0.5434329509735107, "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_RANT_reg/4eb46370c3a2206135f4e61d94761250ef48b135b824ab4c2d58a900b2b87d15.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "4ebc47e3fda2c52aea03f57196404cd80ec3f3f6918c1818fd95d4bf1ef88026", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "027a950a08fd849a986828540372ec5d7073c9f2ec4d405715b7a06414ef286b", "generation_seconds_batch": 24.03266090597026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 5.974957466125488, "genuineness": 3.4843525886535645, "identity": 0.6676086187362671, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4ebc47e3fda2c52aea03f57196404cd80ec3f3f6918c1818fd95d4bf1ef88026.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "4f09701c32cecbe0a19f8422159f91f8c57364d6d2e2591e377c18c5806ec114", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b5208b09cb907628be8545c51548d0de0f5686c03d9309859e3c910e584cc7de", "generation_seconds_batch": 20.044148677028716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4398146867752075, "target_z": 0.00439451158800641, "ce": 6.23706579208374, "genuineness": 4.3881025314331055, "identity": 0.3704039752483368, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4f09701c32cecbe0a19f8422159f91f8c57364d6d2e2591e377c18c5806ec114.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "4f36f236ba0def52cb2b3bb2f866f9eca34d7a1487a6b3b0da5c0687f5ab08b2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1445c10976bcec5235eb8b1fdbe3b2ac4e5669750bfd6004398321511c56000d", "generation_seconds_batch": 16.026484497939236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.325006484985352, "target_z": 2.2656249430668955, "ce": 5.613565921783447, "genuineness": 3.5816311836242676, "identity": 0.21847304701805115, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/4f36f236ba0def52cb2b3bb2f866f9eca34d7a1487a6b3b0da5c0687f5ab08b2.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": "4f4d08d293f0d3e73e5394e961f9da606d6c19292ccfa763db77c5b3a18a3654", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12e35a3540623b2d62604dd7a451b0cb358109e362070088c1adc132d3f84833", "generation_seconds_batch": 16.18593834794592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906339168548584, "target_z": 1.9375000408750493, "ce": 5.689021587371826, "genuineness": 3.5941832065582275, "identity": 0.6570460796356201, "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_RANT_reg/4f4d08d293f0d3e73e5394e961f9da606d6c19292ccfa763db77c5b3a18a3654.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": "4f59d96115821f56f6b4364398df44d0948b4fa98b78d1be0618a442b933ac12", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56f574358fbb6b6c215aef9be71543e6d61f99d88be39f81968b650177da4a0f", "generation_seconds_batch": 21.21213507698849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6471638679504395, "target_z": 1.734374870075736, "ce": 6.141753673553467, "genuineness": 3.660081624984741, "identity": 0.6803560256958008, "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_RANT_reg/4f59d96115821f56f6b4364398df44d0948b4fa98b78d1be0618a442b933ac12.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "4fb3dc17100aa15ecb9ef250abbe4e8b9a39e2345647e4f9ed78db323fcb14ac", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2facb4ea5f0032e3bc16aafc60c0a0e3e03f740e72a6cfc9144417bb4d67f0c6", "generation_seconds_batch": 20.93573178409133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5044355392456055, "target_z": 2.406250061312574, "ce": 5.847874641418457, "genuineness": 3.537698984146118, "identity": 0.252289742231369, "hypothesis": "La porte 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_RANT_reg/4fb3dc17100aa15ecb9ef250abbe4e8b9a39e2345647e4f9ed78db323fcb14ac.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "50413b7cfb794eb18375d2ecde8fb678d0b7500b06120b6d671bc6f8172264c2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a723d5b6d7c81dcaa897114a6ddfed95dc51ef9fead2520fc65942136bb49d04", "generation_seconds_batch": 14.801876105950214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4348305463790894, "target_z": 0.00048825830340423293, "ce": 6.046441555023193, "genuineness": 4.080576419830322, "identity": 0.6093102693557739, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/50413b7cfb794eb18375d2ecde8fb678d0b7500b06120b6d671bc6f8172264c2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "50d662075b1c384a12e73ac31d0a4d65d7a0f677d39f7757a52aa15ec69fd4ac", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b6b5d826fcd6bceffcec865b530b1f1a56caac0d6e0a9cfef7d2073f8ab8c19c", "generation_seconds_batch": 18.538346052984707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1226990222930908, "target_z": -0.24414064343026778, "ce": 6.0305562019348145, "genuineness": 3.785602569580078, "identity": 0.5520884990692139, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/50d662075b1c384a12e73ac31d0a4d65d7a0f677d39f7757a52aa15ec69fd4ac.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "522352423e2b37dc54836cfee5469cf9182247c6ef558288cf474716b24ef95d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "21eaf185aa032c5162fcbfcb641ecd0f6da6c1c94fa8a146a4648b00d6009d0b", "generation_seconds_batch": 21.327272180002183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.006021976470947, "target_z": 2.015625106567093, "ce": 5.780362129211426, "genuineness": 3.6663577556610107, "identity": 0.4983914792537689, "hypothesis": "J'ai trouvé la 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_RANT_reg/522352423e2b37dc54836cfee5469cf9182247c6ef558288cf474716b24ef95d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "532a04082f71ede1b9ce949bd5fe330389db15a5e32c83e5adb0555dff472004", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a1b8d08c5efaac97dedd64f33cee5a7a7489dbf880de1b510a947f708be3e182", "generation_seconds_batch": 20.68142329296097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6770687103271484, "target_z": 1.757812389783349, "ce": 5.673136234283447, "genuineness": 3.5063188076019287, "identity": 0.4071190357208252, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/532a04082f71ede1b9ce949bd5fe330389db15a5e32c83e5adb0555dff472004.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "535f3aac89bfa5ddfef95247fb07ce6290398143ea2c6926a8e17c5d48f94579", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e08a62c61b6f00b6b6094be7e05d17df6dd5d542af5e3552054ffdef9f832362", "generation_seconds_batch": 24.125085332081653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1202069520950317, "target_z": -0.24609377007256886, "ce": 6.284721851348877, "genuineness": 4.118232727050781, "identity": 0.6709358096122742, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/535f3aac89bfa5ddfef95247fb07ce6290398143ea2c6926a8e17c5d48f94579.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "5368b5ab9192ce26112fa938eabf8e9e0c9cb366b856dbae32489adb3a7f9dab", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2b206fe5005136d0b21713d938f7331f1da77b94ed682f92d09cae09e81b46e7", "generation_seconds_batch": 22.36113321292214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4759496450424194, "target_z": 0.032714801187030115, "ce": 6.006728172302246, "genuineness": 3.71028995513916, "identity": 0.39296334981918335, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5368b5ab9192ce26112fa938eabf8e9e0c9cb366b856dbae32489adb3a7f9dab.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "5391f986b791bc034d76fe43a12d92677a87448ef3db4827802111a2d6f3639c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "38d8ac7c14589f456684f1ed3a9741d11bbb9418ee0522b411a3db70b9b0d54d", "generation_seconds_batch": 24.72982134995982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.407925605773926, "target_z": 1.5468750861295681, "ce": 6.0305562019348145, "genuineness": 3.3588318824768066, "identity": 0.6772746443748474, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5391f986b791bc034d76fe43a12d92677a87448ef3db4827802111a2d6f3639c.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": "53d41bad5d10bda560634a9f4f4cfc8ec9e4f0be5516c5d73163446f04613c4b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2303b1fd7b4acfc4e754bc9de0e6027d409bf28fd41642ea90b6bd31d4ab3517", "generation_seconds_batch": 15.654827943071723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.889575481414795, "target_z": 1.1406251182456784, "ce": 5.458683967590332, "genuineness": 3.735394239425659, "identity": 0.6678611040115356, "hypothesis": "Llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. ¡Llévatelas cuando te vayas!", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/53d41bad5d10bda560634a9f4f4cfc8ec9e4f0be5516c5d73163446f04613c4b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "54512e3de3b8a943f91b50b9f68a5db96b9b3fc21072374e8befb29f07a03195", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9b8d5d306ee64cd1250f49c70345a06bf0c8bdd8f565e656eb729ee8f0a0aacf", "generation_seconds_batch": 15.133480048039928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018712878227233887, "target_z": -1.1093749985401768, "ce": 6.284721851348877, "genuineness": 4.174716949462891, "identity": 0.563595175743103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/54512e3de3b8a943f91b50b9f68a5db96b9b3fc21072374e8befb29f07a03195.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "47b855939d2e1f1dadf28d7ee30f7ae7e9ee14b6aae4223c1bf974d24550a37c", "generation_seconds_batch": 15.127355437958613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13833212852478027, "target_z": -1.0156250131384088, "ce": 6.221180438995361, "genuineness": 4.243753433227539, "identity": 0.6619528532028198, "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_RANT_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "54f0aba89ec3c027834a6e2544031f3049caed74bc0ea1c3f3a44dda2f5c3d59", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a9cf6d38727fa430986ee9f546afe94fb330a7d98a60b6849585f161366f0858", "generation_seconds_batch": 23.661306763999164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7282716035842896, "target_z": 0.23046875693416016, "ce": 5.974957466125488, "genuineness": 3.9048471450805664, "identity": 0.5793536901473999, "hypothesis": "La porte 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_RANT_reg/54f0aba89ec3c027834a6e2544031f3049caed74bc0ea1c3f3a44dda2f5c3d59.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "553c9ea7b0e2d4caff0a72dedc8ae3dc5701ef316f3eb52fab950a8636bf6577", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9946ff0cf0055d03ac08b130941117f40c23fc5ce0055afd7e27ed28cb9c3d5d", "generation_seconds_batch": 17.217230813927017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1077466011047363, "target_z": -0.2558594032840743, "ce": 6.197352409362793, "genuineness": 4.193545341491699, "identity": 0.6617067456245422, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/553c9ea7b0e2d4caff0a72dedc8ae3dc5701ef316f3eb52fab950a8636bf6577.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "55a277917f774fe926f50b795a1301e3e86d8b2de388bcfde937d23cd863774d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1431b556149612ddc97bec2d9a553f70ef558b9293915ed66c5b6a133832d208", "generation_seconds_batch": 24.25824624102097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.799860954284668, "target_z": 1.0703125591228393, "ce": 5.824047088623047, "genuineness": 3.8295347690582275, "identity": 0.43341901898384094, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/55a277917f774fe926f50b795a1301e3e86d8b2de388bcfde937d23cd863774d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "56b2cdb832cfc103464d0737cc64507fccfb8f98d0240e58fef65eb8e971984d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d346431b489e1d2854835e2b1b9885bc8919691d98bd65eb28f5f41d736fa7f9", "generation_seconds_batch": 22.60050759103615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0572245121002197, "target_z": 0.4882811934318514, "ce": 5.911416053771973, "genuineness": 3.886019229888916, "identity": 0.635017991065979, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la cabeza.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/56b2cdb832cfc103464d0737cc64507fccfb8f98d0240e58fef65eb8e971984d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "56f4475dde5580af021eefc00d8b7c1424d34c2ade23937a52491400d7e4d6f2", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b7daeec780bc08f5220f3a30f4ea6e886ef167e5bc3ddf945af7ee5efbb2e7d9", "generation_seconds_batch": 18.33736811601557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.746962070465088, "target_z": 0.24511716003707623, "ce": 6.157639026641846, "genuineness": 4.143336772918701, "identity": 0.3256498873233795, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/56f4475dde5580af021eefc00d8b7c1424d34c2ade23937a52491400d7e4d6f2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "57c75c34eb7f6448aa5f830ece4c136d6faefdbb6961bc356f56f11c05d1c216", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "52c7091b5121f79fd71fb230230aa2efa77fab1de962d48574a0583d97fde8cb", "generation_seconds_batch": 22.283399573992938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13833212852478027, "target_z": -1.0156250131384088, "ce": 6.23706579208374, "genuineness": 4.068024158477783, "identity": 0.6749796867370605, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/57c75c34eb7f6448aa5f830ece4c136d6faefdbb6961bc356f56f11c05d1c216.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "57e7d8a47299ce9a39007a536f22446ecc934b63b091f03c41083e48f3fb50a9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cefb3021b5b1703b8de5e0cae43ce045b42a8805eedeba9b4d2c7ef73747b6a3", "generation_seconds_batch": 22.15474002296105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8959208726882935, "target_z": -0.42187498102229853, "ce": 6.062326908111572, "genuineness": 3.785602569580078, "identity": 0.609036922454834, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/57e7d8a47299ce9a39007a536f22446ecc934b63b091f03c41083e48f3fb50a9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "5806febdb56332a4cafc45f7e0cf8ef05ff3e9b962277b5fb60dc77c7a308f09", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "204cb25eaaa7c87b16428074fdcf0cb91fb484a546defbe5b3f4ee835009a183", "generation_seconds_batch": 21.490800368017517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3680524826049805, "target_z": 1.5156250598527508, "ce": 5.593709468841553, "genuineness": 3.747946262359619, "identity": 0.4335722029209137, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5806febdb56332a4cafc45f7e0cf8ef05ff3e9b962277b5fb60dc77c7a308f09.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "582a38e2c52cf64a5f524dafcd73eb3fc5179ad19f5cb3a9b0621e6c758061f7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5c70a6fcce7d57091547a93c1f5c71c8e406bec29150e8e39b7599785c8e832d", "generation_seconds_batch": 14.801876105950214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5656065940856934, "target_z": 0.8867188416039051, "ce": 6.070269584655762, "genuineness": 3.9801597595214844, "identity": 0.6734215617179871, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/582a38e2c52cf64a5f524dafcd73eb3fc5179ad19f5cb3a9b0621e6c758061f7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "584f12dde87d59620997847831deda36392ea7320bb610c0e5deb241342628e2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b7637d548d7902c95c29c03568b5f065609c359eb4f5c62e3092e5668a664520", "generation_seconds_batch": 22.32018464396242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3133422136306763, "target_z": -0.09472659543726072, "ce": 6.23706579208374, "genuineness": 4.124508857727051, "identity": 0.6873109340667725, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/584f12dde87d59620997847831deda36392ea7320bb610c0e5deb241342628e2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "587c7eb82ade8be6bc75baeb2ca2a099fb4d2761ef7ebffebd97017f668c5b06", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9024a9a34331fc53fb8b78c83bcea21adb8dbe5b68fabbb96fb818464c076d5", "generation_seconds_batch": 22.434583211899735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7500195503234863, "target_z": 1.0312500262768174, "ce": 5.434855937957764, "genuineness": 3.653805732727051, "identity": 0.6747544407844543, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/587c7eb82ade8be6bc75baeb2ca2a099fb4d2761ef7ebffebd97017f668c5b06.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "588aa633cc54543904fee5d261b170cea8eff1abe65888a4616a59b1eab1a78b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c7f4b744d9ad46d764245faa2e24455d754c245cc03c235557838228e007f8e3", "generation_seconds_batch": 23.301011241041124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3027509450912476, "target_z": -0.10302736030986931, "ce": 6.332377910614014, "genuineness": 4.645420074462891, "identity": 0.4509916305541992, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/588aa633cc54543904fee5d261b170cea8eff1abe65888a4616a59b1eab1a78b.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": "589426477685f72854ac19cce31edb48550dadf8f38ab18cd5f018bf3b2ab9e3", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a7e4658e81313a4cb6bbf128024e1d80d742607398d5c4b135b28866d32f1d8c", "generation_seconds_batch": 16.295550673035905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6004953384399414, "target_z": 0.9140624277387521, "ce": 5.7247633934021, "genuineness": 3.5439748764038086, "identity": 0.35802194476127625, "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_RANT_reg/589426477685f72854ac19cce31edb48550dadf8f38ab18cd5f018bf3b2ab9e3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "58a32559c7197b65f400c41c5beec263abaa1527b461461700e06520d5aee185", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a28ea66b2ba0008b1e034c1a1a18d066ddd1602d69a9924bf3b61e02854d15ff", "generation_seconds_batch": 21.641978056984954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27290380001068115, "target_z": -0.9101562678828341, "ce": 5.935244083404541, "genuineness": 4.331617832183838, "identity": 0.6899374127388, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/58a32559c7197b65f400c41c5beec263abaa1527b461461700e06520d5aee185.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "58f5539503f83a1d8788e68d865023ccf7187219b8d0d088a5022d18711b953a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1a7e8cd049c5a6bc172ba216d05c5e81f031ad8518026df25febcd141073c290", "generation_seconds_batch": 22.287493029027246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8197975158691406, "target_z": 1.0859375722612479, "ce": 5.657250881195068, "genuineness": 3.6287014484405518, "identity": 0.35931986570358276, "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_RANT_reg/58f5539503f83a1d8788e68d865023ccf7187219b8d0d088a5022d18711b953a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "58f708eabe9b8983b15ff9b0c8dd9f691e9a3846987bc4984d904f3296177404", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0d154afb358a5cb3d7831fa2c5d541a024577b27348d7cf3625f4660515724ff", "generation_seconds_batch": 20.135888652992435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0055718421936035, "target_z": -0.3359375021897348, "ce": 6.062326908111572, "genuineness": 4.030367851257324, "identity": 0.691818118095398, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/58f708eabe9b8983b15ff9b0c8dd9f691e9a3846987bc4984d904f3296177404.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "59619dbd550a3622da4566b0a3a7dc7ba63adc6eee2d0bcb56c404e90f538641", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da251db55f5dcfd8bbff3330e08ba8d1aadc29f7bd786b4da7fdc44b433f8a10", "generation_seconds_batch": 22.675156541983597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0579053163528442, "target_z": -0.2949218427014119, "ce": 6.165581703186035, "genuineness": 4.243753433227539, "identity": 0.6681698560714722, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/59619dbd550a3622da4566b0a3a7dc7ba63adc6eee2d0bcb56c404e90f538641.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": "5983a7ebd3c12266bfed8b37df4e09bbb06ac67f915e23af87693cccde2027b8", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd28782a2273f1ddaf89388d8632008a44464b1f147be1800d27164d4511789c", "generation_seconds_batch": 21.744843691936694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0223355293273926, "target_z": 0.46093742043963615, "ce": 6.070269584655762, "genuineness": 3.6192874908447266, "identity": 0.3910851776599884, "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_RANT_reg/5983a7ebd3c12266bfed8b37df4e09bbb06ac67f915e23af87693cccde2027b8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "59e2070f4945ffb21eef9e31e2f2971a4e495fa888eb900bd40f6a2f5636b30f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8d6f589b129cfdaa20f2d8c8be6b446a6a52780298d96e07f185e51baee3c2eb", "generation_seconds_batch": 21.911967826075852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8265929222106934, "target_z": 1.8749999883214146, "ce": 5.839932441711426, "genuineness": 3.3243136405944824, "identity": 0.653538703918457, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/59e2070f4945ffb21eef9e31e2f2971a4e495fa888eb900bd40f6a2f5636b30f.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": "59f6317f2c139078b0ad5033a797bfb87cff72b561a8a1f98bf940dbc57f46ba", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b113915925cd8324f125e30a3a5781f431367905fd319200572c3901162c540", "generation_seconds_batch": 23.88891100406181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1886234283447266, "target_z": 1.3749999416070724, "ce": 5.788304805755615, "genuineness": 3.7291181087493896, "identity": 0.5604016780853271, "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_RANT_reg/59f6317f2c139078b0ad5033a797bfb87cff72b561a8a1f98bf940dbc57f46ba.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "5a637f6742851aa4e2d98c36a517772d0603afb01526579e11babf16bd3e96f0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "53fae61bf1572376935cd71f241624822b460f6ca72960103e977003ff064dde", "generation_seconds_batch": 22.281048033968545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.589338779449463, "target_z": 0.12158201669738758, "ce": 6.229123115539551, "genuineness": 4.293961524963379, "identity": 0.44197413325309753, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5a637f6742851aa4e2d98c36a517772d0603afb01526579e11babf16bd3e96f0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "5a9a069a91dcf70734b9f81ade805119a5641ce34b6046da19f610feb2646973", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a1bd66c8763f3c9b133c6dce2727260118899015811142af40665d0b8014cbc9", "generation_seconds_batch": 21.537232307018712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.145577430725098, "target_z": 2.1249998248212174, "ce": 5.669164657592773, "genuineness": 3.3117613792419434, "identity": 0.5236524939537048, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5a9a069a91dcf70734b9f81ade805119a5641ce34b6046da19f610feb2646973.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "5aeea2a44c737b90256c6bc45f84fa3eb9c42288b6c6eb3ed4853ffff438127f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7ff32f807abe3997be538412dd10c6157336daebc7727b4326e5b4453adbc9f6", "generation_seconds_batch": 23.784673233982176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4124596118927002, "target_z": -0.8007812693426573, "ce": 5.847874641418457, "genuineness": 4.15588903427124, "identity": 0.6210734844207764, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5aeea2a44c737b90256c6bc45f84fa3eb9c42288b6c6eb3ed4853ffff438127f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "5b226560a463fafc8f671e1d4bc5fa743f8bd55eee86c75361bd11afdd589584", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "63330bef9792ea8c075653065adf97d5d3020504a03c53a01238e69deb03e1b2", "generation_seconds_batch": 22.741616470972076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4444873332977295, "target_z": 0.008056635720906471, "ce": 6.213237762451172, "genuineness": 3.8546390533447266, "identity": 0.24596884846687317, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5b226560a463fafc8f671e1d4bc5fa743f8bd55eee86c75361bd11afdd589584.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": "5b6695df4f9d110b5767cbca19e72308143f5ad4f57837582b70d54dbb4bb4b0", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f86aa9941cd16b974b9ca210374099b1c90a0647f1fb75dae9c6b42fe1910a91", "generation_seconds_batch": 21.137362504028715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.620431900024414, "target_z": 0.9296874408771608, "ce": 6.054384231567383, "genuineness": 3.7667744159698486, "identity": 0.5095292925834656, "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_RANT_reg/5b6695df4f9d110b5767cbca19e72308143f5ad4f57837582b70d54dbb4bb4b0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "5ba21fc2f877328865470feaef610286d37815ce23d04e4446f4687f98d74469", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1481ef0534f9019710190e6e28a240125f467af32ed09370c1838cf8111d435d", "generation_seconds_batch": 17.361394897918217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6684619188308716, "target_z": 0.18359371751893402, "ce": 5.764476776123047, "genuineness": 4.024092197418213, "identity": 0.5584115386009216, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5ba21fc2f877328865470feaef610286d37815ce23d04e4446f4687f98d74469.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "5bf8d3012ae2dff36cf1fe5e24c20a992c089750397f0149fabac7215837622b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c34640e4356328446d7e5782cf0a03136cab180f4c331a3aa720531a3e3f908a", "generation_seconds_batch": 15.112451046006754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7015395164489746, "target_z": -0.5742187656930993, "ce": 5.9431867599487305, "genuineness": 4.2374773025512695, "identity": 0.5651236772537231, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5bf8d3012ae2dff36cf1fe5e24c20a992c089750397f0149fabac7215837622b.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": "5c1f1f3874e0b78e307ee1e819742ab788a8a31de6f630db8e893d3f1a0423f5", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b3a32485a7eebaf70e46753dea9f3c3a99fecd702e0cb2161c290f18062786c1", "generation_seconds_batch": 25.65482977393549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.863293170928955, "target_z": 2.687499924089194, "ce": 6.141753673553467, "genuineness": 3.5220088958740234, "identity": 0.40117380023002625, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5c1f1f3874e0b78e307ee1e819742ab788a8a31de6f630db8e893d3f1a0423f5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "5c49259e43d8849a56c494f50b29be1e9828c6b44497eed39d46d16e1f2cbea8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "618d333935f8f780acf8e090a50778743d20a84a0006b3fefdcbaa15b2b457db", "generation_seconds_batch": 21.568647903972305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0871293544769287, "target_z": 0.5117187131394645, "ce": 5.967014789581299, "genuineness": 3.8797430992126465, "identity": 0.6685589551925659, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5c49259e43d8849a56c494f50b29be1e9828c6b44497eed39d46d16e1f2cbea8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "5c54f47e0a205076c6ce749037708f9fe8fcb3e8eef2f3ee03773899cec1fcc3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c66a201fc46a9f95291bd79097a43e56cabcd847fda1cbea8b456f6f12ed39e0", "generation_seconds_batch": 23.069238126976416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2185282707214355, "target_z": 1.3984374613146855, "ce": 6.038498878479004, "genuineness": 3.8420867919921875, "identity": 0.6511321067810059, "hypothesis": "J'ai trouvé la 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_RANT_reg/5c54f47e0a205076c6ce749037708f9fe8fcb3e8eef2f3ee03773899cec1fcc3.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": "5c59ded6ea496c2a9d17bab52e577fa868e5ffa88494a2510d311b2cf3443a37", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14d6edaa0f5f1aa3d936afac173253b22cb29124c77c2a51850a5ae624016451", "generation_seconds_batch": 20.729951173067093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9226529598236084, "target_z": 0.3828125416049609, "ce": 6.006728172302246, "genuineness": 3.986435890197754, "identity": 0.46131494641304016, "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_RANT_reg/5c59ded6ea496c2a9d17bab52e577fa868e5ffa88494a2510d311b2cf3443a37.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "5c77666fcd330d0708263c27f3fe28398402493346ba59d630abde0d43148c4a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70591bff1a6a2efd68f600e5c5e112456ef5d98429e9184ab071ac746134f728", "generation_seconds_batch": 20.993413344956934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.607290744781494, "target_z": 1.7031248437989186, "ce": 5.442798614501953, "genuineness": 3.4310061931610107, "identity": 0.3653334379196167, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5c77666fcd330d0708263c27f3fe28398402493346ba59d630abde0d43148c4a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "5cb9fe794a167b01ebdda1d6e9b6e50252b98b45d5cdb7818895c35cbb50eb44", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0fb823f0f2d8e5fa7c38987e3132e170b795568c449c808c5474eee5571c9ec8", "generation_seconds_batch": 24.2149029240245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6172590255737305, "target_z": 1.710937350368123, "ce": 5.387199878692627, "genuineness": 3.6192874908447266, "identity": 0.18694989383220673, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5cb9fe794a167b01ebdda1d6e9b6e50252b98b45d5cdb7818895c35cbb50eb44.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "5cd38e6745302fe043f95099fc83bdb8e3e74ecff3084ed643c93f15d3ed5c1d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "38d4271d25e31cad02b259e526841526e46b702818152432bd54d7a47a717c6e", "generation_seconds_batch": 21.314187624957412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864943981170654, "target_z": -1.093749985401768, "ce": 6.09409761428833, "genuineness": 4.444586753845215, "identity": 0.7413568496704102, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5cd38e6745302fe043f95099fc83bdb8e3e74ecff3084ed643c93f15d3ed5c1d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "5cdd4df5bb2d04072410af766c7fa6f9614579be5fa715c7c9c7968043049b96", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "25e042d35f38b31d57a42d421cc77d790e994b600d988b75a0753829f6df263a", "generation_seconds_batch": 20.135888652992435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2934057712554932, "target_z": -0.11035151514698528, "ce": 6.109982967376709, "genuineness": 4.15588903427124, "identity": 0.6444453597068787, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5cdd4df5bb2d04072410af766c7fa6f9614579be5fa715c7c9c7968043049b96.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "5cf9600f4b5764281daa4c975f47ff7f0f8274c723ebfa20452697409993b102", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4775ac032e438a4485472eab55816ade9dbb06bf884e5c116bf43bff48ec482f", "generation_seconds_batch": 21.57553256303072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5071582198143005, "target_z": -0.726562503649558, "ce": 6.109982967376709, "genuineness": 4.406930446624756, "identity": 0.5934138298034668, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5cf9600f4b5764281daa4c975f47ff7f0f8274c723ebfa20452697409993b102.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "5e074e1ce429ccf647d29ff2b6eb0054e8ad15abcfa7ec5e0e500cc70b48bd84", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d01e1927163eb81ec5cb2c2f6de941ace25a842264a994e2e6eeb3f9bc19d251", "generation_seconds_batch": 23.51941510697361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11839556694030762, "target_z": -1.0312500262768174, "ce": 6.197352409362793, "genuineness": 4.356721878051758, "identity": 0.4577924311161041, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5e074e1ce429ccf647d29ff2b6eb0054e8ad15abcfa7ec5e0e500cc70b48bd84.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "fb78964d526dcdf1ba0df65f7f62076502037c951eb8694dbf293e3fc5df05e5", "generation_seconds_batch": 14.691593277966604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3975071907043457, "target_z": -0.8125000291964638, "ce": 6.141753673553467, "genuineness": 4.400654315948486, "identity": 0.8527224063873291, "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_RANT_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "5ebabc42a58229c638a4989d4072ee730a4075de6ac8768919a69c8667139120", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d8a63751f6df4e86a9dd27220ec5d521b7de522e8b1be3a276ed2ec23e7223b3", "generation_seconds_batch": 25.832968196016736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7898926734924316, "target_z": 1.0625000525536348, "ce": 5.855817794799805, "genuineness": 3.8295347690582275, "identity": 0.6275171637535095, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5ebabc42a58229c638a4989d4072ee730a4075de6ac8768919a69c8667139120.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "5ebc550a28714de87d8204472f174ce1d60eadb825fbfc197e95c540735aab22", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0dbb6012d4e26a6765310eef9fef8751d51e3f37137001622e2bab5dd45c70e2", "generation_seconds_batch": 23.18315552000422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.577385902404785, "target_z": 1.6796873240913055, "ce": 5.133034706115723, "genuineness": 3.292933464050293, "identity": 0.3787984549999237, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5ebc550a28714de87d8204472f174ce1d60eadb825fbfc197e95c540735aab22.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "5efa22256f4511bd812d69aadb57b792d0975c7cb735902137f41a13948b9f90", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "476720b1f888495c18aaa92cc75ae239eb78e23c021f6cf08c1b0ad17c1c3f5e", "generation_seconds_batch": 23.55741221399512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.014859437942505, "target_z": 0.45507813394141705, "ce": 6.09409761428833, "genuineness": 4.042920112609863, "identity": 0.739022433757782, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5efa22256f4511bd812d69aadb57b792d0975c7cb735902137f41a13948b9f90.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "5f3e646fa803c4b88e7f81bbb0b77dd9afbca1843c6ce6c3a7aed4b19cc44bee", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "bf201c8928b312057079371348cd8a310530a6b943dd6128ec5c42c5c6fffc58", "generation_seconds_batch": 20.616082031978294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0703656673431396, "target_z": -0.2851562094899065, "ce": 6.395919322967529, "genuineness": 4.406930446624756, "identity": 0.5656164288520813, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5f3e646fa803c4b88e7f81bbb0b77dd9afbca1843c6ce6c3a7aed4b19cc44bee.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "5fbc759f5775a9acc0f87a34963b8486a305b1763bb3538952f595eeb3f8607f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "0a63637b598d1eea90536b48e720faf2b19b6dbb3d3cffa7e97887af1bd36f49", "generation_seconds_batch": 19.76697251398582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8884446620941162, "target_z": -0.4277343609492018, "ce": 6.268836498260498, "genuineness": 4.068024158477783, "identity": 0.2625487148761749, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5fbc759f5775a9acc0f87a34963b8486a305b1763bb3538952f595eeb3f8607f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "5ffdd79af271e85a9e2d7299623e0d114c7685606bf7dce85a7a6b5051c6c2aa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e9f25dd621e0c9c9ef8e03c51edebcb9a5c92c36c0d357757e6303166a810026", "generation_seconds_batch": 23.069238126976416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9095120429992676, "target_z": 1.1562501313840872, "ce": 5.9431867599487305, "genuineness": 3.8169827461242676, "identity": 0.7995923757553101, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/5ffdd79af271e85a9e2d7299623e0d114c7685606bf7dce85a7a6b5051c6c2aa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "6009d72b36824f411bd4f394fd549e003a5fb171c00e0c8ded5320c21d7bcbfb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63d0f4f1b3022d19ea9cc81520352366e1a5601ab72ee6d90e8175b35ba9d700", "generation_seconds_batch": 24.03266090597026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.198591709136963, "target_z": 1.3828124481762767, "ce": 5.677107334136963, "genuineness": 3.653805732727051, "identity": 0.5810297727584839, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6009d72b36824f411bd4f394fd549e003a5fb171c00e0c8ded5320c21d7bcbfb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "6041dc14b62b0f305663b28a8f88880f186c01de566e8c56224dc46216216943", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "db013fb042bc826c4dc9fc20f381a5cf07e1fd6bf1684df3adbf18993d6addf1", "generation_seconds_batch": 24.33940891502425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0889406204223633, "target_z": 1.2968748759150288, "ce": 5.863759994506836, "genuineness": 3.8671910762786865, "identity": 0.2608909606933594, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6041dc14b62b0f305663b28a8f88880f186c01de566e8c56224dc46216216943.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "60cf57ed8cc1bc9d6d5384d27b92fa257eb88686678e42ceca1ef4442d190033", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "58507045dab43a7ee2e49f87cf918d9b40e3c606de104e9904cc6ac3d27e89a6", "generation_seconds_batch": 21.732938540983014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5500887632369995, "target_z": 0.09082034215265855, "ce": 6.165581703186035, "genuineness": 3.9048471450805664, "identity": 0.48557958006858826, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/60cf57ed8cc1bc9d6d5384d27b92fa257eb88686678e42ceca1ef4442d190033.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "60e0ec52b4f27fdaa6207afd8a6079bd46c38979f118fe89c6577bd7b60c4c5b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3631562c823356669a1396b79c4e658a4eb2a69dc375f4b04dfbaa94314a5de8", "generation_seconds_batch": 24.918112954008393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.208559989929199, "target_z": 1.3906249547454812, "ce": 6.006728172302246, "genuineness": 3.860914945602417, "identity": 0.709140956401825, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/60e0ec52b4f27fdaa6207afd8a6079bd46c38979f118fe89c6577bd7b60c4c5b.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": "60e1978633f749a0659a595323787aab73f5e48154aa8d1cb9bd50749c30ad21", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "944540a0a5209bdb3b109597043478ed087feacd41c5aac45fc478f4b268d399", "generation_seconds_batch": 16.277613817946985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460939645767212, "target_z": 0.8046875226272594, "ce": 5.879645347595215, "genuineness": 3.47493839263916, "identity": 0.4890998601913452, "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_RANT_reg/60e1978633f749a0659a595323787aab73f5e48154aa8d1cb9bd50749c30ad21.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "60efa14114d0f94d46041f1c30b067355e761281647cafe9b0ef92507138f51b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "08fb7f301588a4362933beef7a9cb2ac2d7fc1c247519070464fb2eaa850d953", "generation_seconds_batch": 17.666048213024624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9095120429992676, "target_z": 1.1562501313840872, "ce": 6.070269584655762, "genuineness": 3.7416701316833496, "identity": 0.6434396505355835, "hypothesis": "La porte 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_RANT_reg/60efa14114d0f94d46041f1c30b067355e761281647cafe9b0ef92507138f51b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "610a5cde534af6c9e1336cfa90f2239dfec2c6d31d75aad68d612e3751cbf017", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4850c9b4ea0ddc7e5423b22c0d512c266536610f12bed65f70b6bfc00c69fbee", "generation_seconds_batch": 26.150586207048036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.241637706756592, "target_z": 0.632812564962132, "ce": 5.9431867599487305, "genuineness": 3.923675537109375, "identity": 0.5192460417747498, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/610a5cde534af6c9e1336cfa90f2239dfec2c6d31d75aad68d612e3751cbf017.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": "61f6af8ea399ef85f018b0b13823038d5329190ed51dc9240f846f8e664bbd6c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ef7e575ee8b1adc43f580cd0eef2d4e78298327292b5ab92fe258d1a2c00359", "generation_seconds_batch": 20.714295544079505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 5.839932441711426, "genuineness": 3.5847692489624023, "identity": 0.6082202196121216, "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_RANT_reg/61f6af8ea399ef85f018b0b13823038d5329190ed51dc9240f846f8e664bbd6c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "6224bafe8d5f76d36bdb296d1440feb2b74c47234d03c042d3f2a9b3058203f1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1afe794bb4184521b9ff42fb026fa64b78ec04bbe74e37219614734344e0d020", "generation_seconds_batch": 21.88931135996245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.168686866760254, "target_z": 1.3593749284686636, "ce": 5.77639102935791, "genuineness": 3.9425034523010254, "identity": 0.49508875608444214, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6224bafe8d5f76d36bdb296d1440feb2b74c47234d03c042d3f2a9b3058203f1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "62db149cc54a9005e623dc468e21d3f14c542169d14e9bd2d995b364a8715d30", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1653e6d5e311525d7b1cca58fde43b7f6da2fb8e3b91da5098761212b1fc2f01", "generation_seconds_batch": 21.953229532926343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4319491386413574, "target_z": -0.001770006421353462, "ce": 6.165581703186035, "genuineness": 4.086852550506592, "identity": 0.6968134641647339, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/62db149cc54a9005e623dc468e21d3f14c542169d14e9bd2d995b364a8715d30.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "630028fce852fb6d043f1d0a9c780c39e8d8325263afba9b23e0db21d6613e2a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9879ffd0dbb75ce648d25b72697420930878330dd53cdc61352051e49fb91958", "generation_seconds_batch": 23.479194947052747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6304001808166504, "target_z": 0.9374999474463651, "ce": 5.879645347595215, "genuineness": 3.7918784618377686, "identity": 0.5735737085342407, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/630028fce852fb6d043f1d0a9c780c39e8d8325263afba9b23e0db21d6613e2a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "633d07547cff2ffb4debc739df3389f715f92b8ae1c82e6c5f5cf9a7218cb9aa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "37df65c576223258075604088206029c7006dfb6a6c660e064ad6d4891e71507", "generation_seconds_batch": 23.47590358403977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8585398197174072, "target_z": -0.45117188065681485, "ce": 6.149696350097656, "genuineness": 3.873466968536377, "identity": 0.3270176649093628, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/633d07547cff2ffb4debc739df3389f715f92b8ae1c82e6c5f5cf9a7218cb9aa.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": "6427fb0695ef51f150781ce62bfc4c96d1d54823f191456abb1888fb64be51c9", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "088906097ceed90030925b45fff99fb0a8635252b46f653145dfc1ba5d5585be", "generation_seconds_batch": 17.184968113084324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4976401329040527, "target_z": 1.6171876452524074, "ce": 5.895530700683594, "genuineness": 3.760498285293579, "identity": 0.22578157484531403, "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_RANT_reg/6427fb0695ef51f150781ce62bfc4c96d1d54823f191456abb1888fb64be51c9.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": "643bc0bc46a453203a12330ba5c037a8cec2db4b2e92d4a10969bbe714a9c923", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52784162895d3e8280cd169f439b972875fdf8823207a456f27af16904541636", "generation_seconds_batch": 30.151650395942852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460939645767212, "target_z": 0.8046875226272594, "ce": 5.80021858215332, "genuineness": 3.911123275756836, "identity": 0.4891088008880615, "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_RANT_reg/643bc0bc46a453203a12330ba5c037a8cec2db4b2e92d4a10969bbe714a9c923.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": "646cb65e21287be288add7c6b55d79f98f35c7bd807bb7cf9c4c23451a543eeb", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e79147127dceae00fd2e7964975c9cc42a1076eea020814801c24c0dcba15ba1", "generation_seconds_batch": 23.64320979907643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.445421814918518, "target_z": 0.00878902317601282, "ce": 5.879645347595215, "genuineness": 3.7918784618377686, "identity": 0.4127163290977478, "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_RANT_reg/646cb65e21287be288add7c6b55d79f98f35c7bd807bb7cf9c4c23451a543eeb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "64cc50504b41d2abe09136f9505c8b3a6429651935604850e01cd562fa90cda9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "8f6e27db1cd2190d8c63cd5d49d3945a9179c4a4903e89a623272281bd44682f", "generation_seconds_batch": 23.301011241041124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0603973865509033, "target_z": -0.29296871605911085, "ce": 6.157639026641846, "genuineness": 4.400654315948486, "identity": 0.4978141188621521, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/64cc50504b41d2abe09136f9505c8b3a6429651935604850e01cd562fa90cda9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "6518594f5328e4731f8d7f2148b2306224abab34d07f60d1d0c93dc5d7e9bbff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0af69f296103bc4dd659a360d28287b09c43b97c195678557c5903ab121964c7", "generation_seconds_batch": 23.13650703895837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8386032581329346, "target_z": -0.46679689379522354, "ce": 6.062326908111572, "genuineness": 4.293961524963379, "identity": 0.8145931959152222, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6518594f5328e4731f8d7f2148b2306224abab34d07f60d1d0c93dc5d7e9bbff.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": "6633dad43631f68ea24b74d4fc3b03898a7b4569cf7e33ad89cfa36ca707325d", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8da50fed56b7fd2e84a9f3f2670f86d6fd22176dd3d20f373276a663b79817f0", "generation_seconds_batch": 16.280907156062312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7069735527038574, "target_z": 1.7812499094909622, "ce": 5.824047088623047, "genuineness": 3.7542223930358887, "identity": 0.4100521206855774, "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_RANT_reg/6633dad43631f68ea24b74d4fc3b03898a7b4569cf7e33ad89cfa36ca707325d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "665479f9c311cea9b3a7307ff9b67a48caaca48f6ae5d715a44288743e66b397", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a6f802f9e0c68fbef7a8480311197ac4c9e35f62384892ad3a24227a102c9268", "generation_seconds_batch": 23.554372834041715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.525733470916748, "target_z": 0.8554688153270877, "ce": 6.0305562019348145, "genuineness": 3.6475296020507812, "identity": 0.29093706607818604, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/665479f9c311cea9b3a7307ff9b67a48caaca48f6ae5d715a44288743e66b397.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "67121743f7e56a757599e6e92ad9c4fd253d19f9cde869b60ad0175b79f031c0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6822261406628e930c4ae530884c78e192c5acc897924f4bd27521be209fbc41", "generation_seconds_batch": 15.942274330998771, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.597322463989258, "target_z": 1.6953123372297143, "ce": 5.470598220825195, "genuineness": 3.340003728866577, "identity": 0.5248192548751831, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready for-", "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_RANT_reg/67121743f7e56a757599e6e92ad9c4fd253d19f9cde869b60ad0175b79f031c0.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": "671ea0c0a3429ef2c148ca74f2c14877c6a41c33709fd90706b1905bb372bd43", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb80a0c46b715da59e48cd0d5ea3152dd0018faa9f928721263d5d9397369932", "generation_seconds_batch": 22.6342873100657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.989258050918579, "target_z": 1.2187499970803537, "ce": 5.7684478759765625, "genuineness": 3.7416701316833496, "identity": 0.7796942591667175, "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_RANT_reg/671ea0c0a3429ef2c148ca74f2c14877c6a41c33709fd90706b1905bb372bd43.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "67c27b7a56d78c523dfdf3a9e788c81457933b41adc8a91cc5ce12c685c38fa0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1119eb6c1bad0530e687bcac97c1abb2ff0cbdcd406b4a2c5a3baafcec760133", "generation_seconds_batch": 20.974749048938975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.564245223999023, "target_z": 2.4531251007278003, "ce": 6.054384231567383, "genuineness": 3.4372823238372803, "identity": 0.5838396549224854, "hypothesis": "La porte 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_RANT_reg/67c27b7a56d78c523dfdf3a9e788c81457933b41adc8a91cc5ce12c685c38fa0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "67f84ad6e1b269470170e1c052e5ac929997f1b5b19d6d9b095ac90027a4f6b2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d2ca7c14abcf4ceff9b9e1d5fbce93284b486ade84b386e92f664d933b42893b", "generation_seconds_batch": 23.18315552000422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348115921020508, "target_z": 1.5000000467143422, "ce": 5.474569320678711, "genuineness": 3.4937667846679688, "identity": 0.4201633334159851, "hypothesis": "The door is still open! I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/67f84ad6e1b269470170e1c052e5ac929997f1b5b19d6d9b095ac90027a4f6b2.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": "683256b9840eedeadcc9fe61174b5a466a3a0dce61ca45a50a0830476576f911", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e30863c65db15bda2f1a386075439d79ee3afc8731bf54cf7130719cc0f5843b", "generation_seconds_batch": 30.15688671998214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680241584777832, "target_z": 0.9765624802923869, "ce": 6.1894097328186035, "genuineness": 3.7918784618377686, "identity": 0.6642642021179199, "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_RANT_reg/683256b9840eedeadcc9fe61174b5a466a3a0dce61ca45a50a0830476576f911.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "697ab4bbf226409875ca274f3ac2f6c5a3110e167c6597cae054c73379e00eba", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5ac18e1b6320502d6e6a401a7a5050dd01fbc94b31aef7f09d0039c635b81255", "generation_seconds_batch": 23.764851261978038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5656641721725464, "target_z": 0.10302736030986931, "ce": 6.117925643920898, "genuineness": 3.8546390533447266, "identity": 0.5254347920417786, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/697ab4bbf226409875ca274f3ac2f6c5a3110e167c6597cae054c73379e00eba.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "697bf6f193b7918223a1649e938b101c4ed2a98b7aa1a74ef29a05d77fe992e9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5233849c7f5ce4f15d747675b54ed15c77c784a05480b52c02d2448fd7865a69", "generation_seconds_batch": 24.172602790989913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0871293544769287, "target_z": 0.5117187131394645, "ce": 6.054384231567383, "genuineness": 3.8420867919921875, "identity": 0.665252685546875, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/697bf6f193b7918223a1649e938b101c4ed2a98b7aa1a74ef29a05d77fe992e9.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "7adc4a69848241026dc1cc9ad2eb281d7d0722e9d91cf196cf018ea7d9792237", "generation_seconds_batch": 15.08116825891193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.021160244941711426, "target_z": -1.1406250248169942, "ce": 6.260893821716309, "genuineness": 3.8797430992126465, "identity": 0.31868281960487366, "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_RANT_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.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": "6adfeeaa33458bc0814f086f36e0ba6775c287c26b528ce140ebfe3ef1fa21c2", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35449f1c8a036b12221a699c49f44a0ec4b9c16d0caad6b5c528560878ec953f", "generation_seconds_batch": 20.74382221198175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5656065940856934, "target_z": 0.8867188416039051, "ce": 5.665193557739258, "genuineness": 3.7542223930358887, "identity": 0.4856153130531311, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en taarna naast het raam. Neem ze mee als je weggaat!", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6adfeeaa33458bc0814f086f36e0ba6775c287c26b528ce140ebfe3ef1fa21c2.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": "6b003272b84d7066b999ce70bfba647fc62272012e907cc35802527315c72163", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69a852f0ed19c67174fdd088ec3da269f128cbe47c0d63c09a8b1c58613de1ae", "generation_seconds_batch": 23.88891100406181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4110982418060303, "target_z": 0.7656249897812377, "ce": 5.855817794799805, "genuineness": 3.8044304847717285, "identity": 0.6623300313949585, "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_RANT_reg/6b003272b84d7066b999ce70bfba647fc62272012e907cc35802527315c72163.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "6b125f7512ff342ae8816085dfe11e67ac950dd332deb6278ebca0d15f1b2a37", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c55e649bbdda8681bdd69da4923242ad8c7da836893c275e6eeb4c5f79860926", "generation_seconds_batch": 23.35153289604932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8098292350769043, "target_z": 1.0781250656920436, "ce": 5.677107334136963, "genuineness": 3.8420867919921875, "identity": 0.6954042911529541, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6b125f7512ff342ae8816085dfe11e67ac950dd332deb6278ebca0d15f1b2a37.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "6b35b193c51bb80bf01e43bf003f91da83859c99806f6839bedbb7b4e147bcff", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b73aa113c27c0e48801054ab96f72c09d400ced4800c6e9e8bdcbaa22193ed1b", "generation_seconds_batch": 17.217230813927017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1650642156600952, "target_z": -0.21093749051114927, "ce": 6.157639026641846, "genuineness": 4.093128204345703, "identity": 0.6840425729751587, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6b35b193c51bb80bf01e43bf003f91da83859c99806f6839bedbb7b4e147bcff.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "65b335ddf697b7c8ec9c11ac3e8d0cee72ce04f424f5eb5c2b640d4c66dda2bf", "generation_seconds_batch": 16.60622680198867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16823697090148926, "target_z": -0.9921874934307956, "ce": 6.149696350097656, "genuineness": 4.607763767242432, "identity": 0.6665632724761963, "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_RANT_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "6bb732e2e269d6bf5bf680029dbeb587c547036dde80397447611892e551a983", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d5f7522a75ebdd3fa151a2b259c08b1e876ea8adb045ba01b9dc9b22482f6f94", "generation_seconds_batch": 16.971062553930096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5270947813987732, "target_z": -0.7109374905111493, "ce": 6.149696350097656, "genuineness": 4.406930446624756, "identity": 0.5137529373168945, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6bb732e2e269d6bf5bf680029dbeb587c547036dde80397447611892e551a983.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": "6bbf412cdfe977104513c6145913c082c6564f94e9e477bf3f5eb066e6d28c97", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bc2982f76f79bfc1548b7701710e196c84219889685549714917547ff7bf89c", "generation_seconds_batch": 21.256950397975743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.95935320854187, "target_z": 1.1953124773727406, "ce": 6.0305562019348145, "genuineness": 3.7793264389038086, "identity": 0.2247045934200287, "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_RANT_reg/6bbf412cdfe977104513c6145913c082c6564f94e9e477bf3f5eb066e6d28c97.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": "6c2d132430a334f6ee8dca529593764297c85692317c9a50bce63f5e8e094cc6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ae68c5fa04ba303a2dfe69ba2d81fc38efdd0538cac583b5b57aac9a9955bee", "generation_seconds_batch": 26.509505647001788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4808762073516846, "target_z": 0.8203125357656682, "ce": 6.141753673553467, "genuineness": 3.6789097785949707, "identity": 0.5698901414871216, "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_RANT_reg/6c2d132430a334f6ee8dca529593764297c85692317c9a50bce63f5e8e094cc6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "6c30705d26eca3b9de2ea7083b59539d2f8ee455189f4332a00ebb2a14281fdd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7e1b2566d6c1a3f1206b2c6495d86ea47487be55f0b8eef8b303615976113fc5", "generation_seconds_batch": 22.821869064006023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.225323677062988, "target_z": 2.187499877374852, "ce": 5.847874641418457, "genuineness": 3.7291181087493896, "identity": 0.36763936281204224, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6c30705d26eca3b9de2ea7083b59539d2f8ee455189f4332a00ebb2a14281fdd.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": "6c38b1df3dcdb7398a74368c3195beb2b2d5f72cc2ffc800a882043a715deec0", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4079b7233e2e000d93e332cf0819b16df6304d13c0b693ed105adf3e1730a59a", "generation_seconds_batch": 26.509505647001788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4110982418060303, "target_z": 0.7656249897812377, "ce": 6.165581703186035, "genuineness": 3.653805732727051, "identity": 0.4677194356918335, "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_RANT_reg/6c38b1df3dcdb7398a74368c3195beb2b2d5f72cc2ffc800a882043a715deec0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "6c99b79bccd824c6027dec3e048a16bfca6b3f1fb48ca360ba04936daca8685c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3492f04ce3e07c63dee4922cefa98064f93c33d311b9bf3c43c519137fa95ce2", "generation_seconds_batch": 23.972582131042145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9693214893341064, "target_z": 1.203124983941945, "ce": 5.967014789581299, "genuineness": 3.5439748764038086, "identity": 0.5284469127655029, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6c99b79bccd824c6027dec3e048a16bfca6b3f1fb48ca360ba04936daca8685c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "6d1d01c1153ece52ed124fba59d106dbd030aeaa0321fb509560b59569cca872", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "9b05f7040c1c5f70395584af1310feb7695d873032253be88b735a6522c0e377", "generation_seconds_batch": 19.969923511031084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35763418674468994, "target_z": -0.8437499620445971, "ce": 6.109982967376709, "genuineness": 3.9676074981689453, "identity": 0.6016120910644531, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "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_RANT_reg/6d1d01c1153ece52ed124fba59d106dbd030aeaa0321fb509560b59569cca872.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "6ddb54a09f50d714f76cd93bd6a7b0282efb8f3dcd769f43d524a503548bdb14", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "171b50b14721cd16e10f41417047cdd3f4ee5d7590b79788c4fa369452ed0d32", "generation_seconds_batch": 21.56533033098094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1712943315505981, "target_z": -0.20605472061973862, "ce": 5.673136234283447, "genuineness": 3.9425034523010254, "identity": 0.6678968071937561, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6ddb54a09f50d714f76cd93bd6a7b0282efb8f3dcd769f43d524a503548bdb14.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": "6dfdd121ac14e24eebaa11f297813d6b21584154bb05b9f2bbfd82cbf23607bb", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "630876ba13fcecfeec1e9b188b9d86eab47bc630d632296a8062e4a46493fef5", "generation_seconds_batch": 20.938480478012934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348115921020508, "target_z": 1.5000000467143422, "ce": 5.5539960861206055, "genuineness": 3.550251007080078, "identity": 0.6228864192962646, "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_RANT_reg/6dfdd121ac14e24eebaa11f297813d6b21584154bb05b9f2bbfd82cbf23607bb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "6e9e5bdbcadcb0273c70e3ae360f4c9939b2683d87e04f35e7ba1e9ccce79b40", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ad8c6b4d6f2ba99a7a158f646fb6828c08490cef0cd88b2d92f8be06b7613639", "generation_seconds_batch": 20.118944074958563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.267919659614563, "target_z": -0.9140625211674362, "ce": 6.284721851348877, "genuineness": 4.21237325668335, "identity": 0.5077543258666992, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6e9e5bdbcadcb0273c70e3ae360f4c9939b2683d87e04f35e7ba1e9ccce79b40.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": "6ec1211b285ffc192e34e59a0554e83a1f20926aba532c95f919ed58f4d50f16", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4a54f2b1e884542cea756655f4238181e8567d2a404f8315450c772db90c9dc", "generation_seconds_batch": 16.398936051991768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.966148853302002, "target_z": 1.9843750802902755, "ce": 5.935244083404541, "genuineness": 3.525146961212158, "identity": 0.5230076909065247, "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_RANT_reg/6ec1211b285ffc192e34e59a0554e83a1f20926aba532c95f919ed58f4d50f16.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "6ee4b1862129946870254e21c2726f586338662cae49b1d26859cd17fca05a49", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7c6918508735a07b4aa74f1452b897c131c8d8b29a6bd09fb2cd6a800ee75647", "generation_seconds_batch": 15.238079856033437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0778417587280273, "target_z": -0.27929692299168735, "ce": 5.804190158843994, "genuineness": 3.9173994064331055, "identity": 0.511103630065918, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/6ee4b1862129946870254e21c2726f586338662cae49b1d26859cd17fca05a49.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": "6ffd8543695842cd8d723a7aee984c5224fd726d4c139cbb72014ab884ae437b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63ac6ac9e237f22b6b0f90fcc19437851a6cd2320f1e601b82541544154cd719", "generation_seconds_batch": 16.277613817946985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4378304481506348, "target_z": 1.5703126058371812, "ce": 5.935244083404541, "genuineness": 3.478076457977295, "identity": 0.4412184953689575, "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_RANT_reg/6ffd8543695842cd8d723a7aee984c5224fd726d4c139cbb72014ab884ae437b.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": "70180fcef5f059b1711831f8a2b07ba4bb1e8dd9eafdad43ef7061bf8314b7e0", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "1b6b8a31e4b62314ffbdac84e417289cb154f2c6f4c747752441ba581e819ef1", "generation_seconds_batch": 21.067508641048335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7898926734924316, "target_z": 1.0625000525536348, "ce": 6.062326908111572, "genuineness": 3.550251007080078, "identity": 0.5066059231758118, "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": 0.120000000000001, "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_RANT_reg/70180fcef5f059b1711831f8a2b07ba4bb1e8dd9eafdad43ef7061bf8314b7e0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "70209567d1858176f6148ac2fcbd038ce46a9479a6809b7a0e7a8fdaf3add4b1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e95dffefdee71786e9649bf6baddab9151f1e2d638587d84bd205fe7f9816531", "generation_seconds_batch": 25.67660490097478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.222381830215454, "target_z": -0.16601557773822423, "ce": 6.149696350097656, "genuineness": 3.8420867919921875, "identity": 0.5157759189605713, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/70209567d1858176f6148ac2fcbd038ce46a9479a6809b7a0e7a8fdaf3add4b1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "709fc3fd59d64caf68e449e35591cce915682464634db23fa5a3d188c9f5d9a2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "238908dbca84ac7a5f37015f1d50f6bda691ce1b80f8ad3a343fd7aa1631c22c", "generation_seconds_batch": 20.01793949108105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2484331130981445, "target_z": 1.4218749810222986, "ce": 6.197352409362793, "genuineness": 3.9801597595214844, "identity": 0.46263670921325684, "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_RANT_reg/709fc3fd59d64caf68e449e35591cce915682464634db23fa5a3d188c9f5d9a2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "70d76cdf2cadac63267ac84ee853f257f7054227ce5b075d935720df5c87189a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4d07b38ca21cdfed41faf071372655fd5a263d55d1c0d8c9744e60c0e2ed83de", "generation_seconds_batch": 20.58774489501957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.287798523902893, "target_z": -0.11474612016367584, "ce": 6.260893821716309, "genuineness": 3.9676074981689453, "identity": 0.27408429980278015, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/70d76cdf2cadac63267ac84ee853f257f7054227ce5b075d935720df5c87189a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "71089a51481e0e1c96fbc77e6ff426aa15257f394488ee5a48ebebb1737a4dc2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2eb03a76e14ab9735b4d15ac6c57f1bc72161a7fe77ea1f5648d3688f1fd1da1", "generation_seconds_batch": 20.809345219051465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.751946210861206, "target_z": 0.2490234133216784, "ce": 6.213237762451172, "genuineness": 3.8671910762786865, "identity": 0.648032009601593, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/71089a51481e0e1c96fbc77e6ff426aa15257f394488ee5a48ebebb1737a4dc2.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": "71b5b5bc4253a82e07ee3b16953b03a26ebd47cb7f81f2af097d065bc6b5aff1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f25ae7c67e06b5ec484d291dd3cd5ca6cdc25619c7e57b53484196caa762e1b0", "generation_seconds_batch": 16.260466612991877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9262757301330566, "target_z": 1.953125054013458, "ce": 5.831989288330078, "genuineness": 3.6349775791168213, "identity": 0.2407587766647339, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/71b5b5bc4253a82e07ee3b16953b03a26ebd47cb7f81f2af097d065bc6b5aff1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "71f8180757531984bf84309b7c0c55a3eefb0cdafbe09cb176c19c7682b75322", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f34a789d182d968a555f7144e4c95386e79d1d233a93874a2b2ad7a34d8e5155", "generation_seconds_batch": 22.570572588942014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.919480323791504, "target_z": 1.1640626379532915, "ce": 5.808161735534668, "genuineness": 3.600459337234497, "identity": 0.5594319105148315, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/71f8180757531984bf84309b7c0c55a3eefb0cdafbe09cb176c19c7682b75322.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "7260fad599a35227736c51dcb4a535a20139c2ae84a432ec5ab418b39e2deb68", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7336dede35cce526fb2d0e5ad91ca7c642db1554163f1f63a33257bc13d48fc4", "generation_seconds_batch": 20.486734170001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8710001707077026, "target_z": -0.44140624744530943, "ce": 6.300607204437256, "genuineness": 4.055472373962402, "identity": 0.20057155191898346, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7260fad599a35227736c51dcb4a535a20139c2ae84a432ec5ab418b39e2deb68.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": "7314b57f825a98d3d67461d2aed595f1cc9da76143c7809e3de64d4d803580d1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4486bc8204539d3defe35b0f469194723afdda6da4d4e9c7635e74d442f135ce", "generation_seconds_batch": 23.236986513016745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5456700325012207, "target_z": 0.8710938284654964, "ce": 6.133810997009277, "genuineness": 3.923675537109375, "identity": 0.5481556057929993, "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_RANT_reg/7314b57f825a98d3d67461d2aed595f1cc9da76143c7809e3de64d4d803580d1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "73bd360210021fbdafc39d7c6d72bdf7106b7dac2811fbde2861054fcb8187ee", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6f8fdb890af170e7cfae84d346772363a8e22ee0c26205bfff51d18becda91c0", "generation_seconds_batch": 22.234727873001248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7912539839744568, "target_z": -0.5039062532846021, "ce": 6.3641486167907715, "genuineness": 4.055472373962402, "identity": 0.5108609199523926, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/73bd360210021fbdafc39d7c6d72bdf7106b7dac2811fbde2861054fcb8187ee.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": "745efa30fd0bd6b3eccfa1489a2b619fe9e6bfa9efdc4463468adea2450c99d5", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7606b54f4b084858f074f0b9430dd4f142a25914ca57dbfab137745e1cf10103", "generation_seconds_batch": 26.37386155000422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1587185859680176, "target_z": 1.3515624218994593, "ce": 5.847874641418457, "genuineness": 3.6663577556610107, "identity": 0.5014548301696777, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor! Ich warte auf der anderen Seite!", "wer": 0.09523809523809523, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/745efa30fd0bd6b3eccfa1489a2b619fe9e6bfa9efdc4463468adea2450c99d5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "749e694ec8187e2e0a9b38f1feff4cbe607386199ac058303d21cd3e191179ff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ab2867a42b240d32c13b109da584ced06122b23174d3e8bb34b20b45c41e129", "generation_seconds_batch": 19.718365306034684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.846529483795166, "target_z": 1.8906250014598232, "ce": 5.446769714355469, "genuineness": 3.760498285293579, "identity": -0.17684614658355713, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/749e694ec8187e2e0a9b38f1feff4cbe607386199ac058303d21cd3e191179ff.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "74dde7a0fa1ed8fa1e7992ca2455815eb1b4ca5c0b7d4093faad96c642edd856", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e2ac83a73af0fcdd5b0cd509f72b6ae992c99bb278050e276c49d3c53aa91de9", "generation_seconds_batch": 21.884859057026915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.708335041999817, "target_z": 0.21484374379575144, "ce": 6.125868320465088, "genuineness": 3.9048471450805664, "identity": 0.6652973890304565, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/74dde7a0fa1ed8fa1e7992ca2455815eb1b4ca5c0b7d4093faad96c642edd856.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "74ed298ee3beb7afe0612d77bafc995e78946bb45b4cc4f85fef54d6712a8493", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "66e2c7f432b1547fefb9d948ac4a39893d928005181341b56d1b7e06aba49c56", "generation_seconds_batch": 16.36142500001006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.211732864379883, "target_z": 0.6093750452545189, "ce": 5.982900142669678, "genuineness": 3.735394239425659, "identity": 0.5993776917457581, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/74ed298ee3beb7afe0612d77bafc995e78946bb45b4cc4f85fef54d6712a8493.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "74f3b1398904e3fda7cfe04ae970acf238f7ca54c1a1abb8f1b8c31e4f6a8f2e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f928961da7b4096b8524426d953447089b17d30796e899c7dd51bbedd9bf757e", "generation_seconds_batch": 25.801488447003067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6403684616088867, "target_z": 0.9453124540155695, "ce": 5.744620323181152, "genuineness": 3.923675537109375, "identity": 0.31908416748046875, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/74f3b1398904e3fda7cfe04ae970acf238f7ca54c1a1abb8f1b8c31e4f6a8f2e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "754f837faa1fd3e7b70ad367d4bbadd27f2b0213c414f7bf5985b1148c377439", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "042812a43ec72f3dbbd76bdbd1cd097d5b0d50f8ca89db4b0e7ac6945c7cf4d3", "generation_seconds_batch": 16.458508905023336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.031128525733947754, "target_z": -1.1484375313861985, "ce": 5.696964263916016, "genuineness": 4.21237325668335, "identity": 0.5709071159362793, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/754f837faa1fd3e7b70ad367d4bbadd27f2b0213c414f7bf5985b1148c377439.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "75921a6f477dbaeccc5e9f405790c4b9b29001bf57bd97d436c37437ab6ac220", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "912580f94079c2b35891dade53ddc147ac8eb8b4cca2ee43fcdc6ad0e8e3d04f", "generation_seconds_batch": 22.206857913988642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.683414340019226, "target_z": 0.19531247737274054, "ce": 6.221180438995361, "genuineness": 3.9048471450805664, "identity": 0.5972281098365784, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/75921a6f477dbaeccc5e9f405790c4b9b29001bf57bd97d436c37437ab6ac220.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "766b6f411175af86ff301eec06df0c1e099d433303f7636b1563be54d8d30e46", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "92e49cb99ae1f272f6b5313829c397d09198e769110df31bd1f01e2ff6e77a81", "generation_seconds_batch": 24.918112954008393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.705162286758423, "target_z": 0.9960937467153979, "ce": 5.879645347595215, "genuineness": 4.017816066741943, "identity": 0.6523095369338989, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/766b6f411175af86ff301eec06df0c1e099d433303f7636b1563be54d8d30e46.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "7672116d4fbaf94f1b60e4d0ffa3f09cddda5c8a5ee9630fdc46245cdac17415", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d304d0d49b7967aba67ca36f4f44c0f7b0ccbefb25b275936e743a07f2b19b27", "generation_seconds_batch": 24.296484290040098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.937605381011963, "target_z": 0.39453130145876747, "ce": 6.078212261199951, "genuineness": 3.860914945602417, "identity": 0.3899676501750946, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7672116d4fbaf94f1b60e4d0ffa3f09cddda5c8a5ee9630fdc46245cdac17415.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": "768dc56673efe27e40c75fe7f9915f35b53bf35adb8cb18231575d1677b5bc7d", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72c6e269fc89e220ed5dd2ee352dd6ea2f23c28c3d3b30cac4e6af94556d91e6", "generation_seconds_batch": 23.024942449992523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2185282707214355, "target_z": 1.3984374613146855, "ce": 6.1020402908325195, "genuineness": 3.660081624984741, "identity": 0.7064564228057861, "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_RANT_reg/768dc56673efe27e40c75fe7f9915f35b53bf35adb8cb18231575d1677b5bc7d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "76d29a90032b3ec5b70d378a54fcb295040c9d42024136f939eada7f44c28ca9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "277a2a3b078483c85c3962ed42a0f85e2210d61ae464a110460109dc6da573f9", "generation_seconds_batch": 22.184799488983117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.387989044189453, "target_z": 1.5312500729911596, "ce": 5.80021858215332, "genuineness": 3.5471129417419434, "identity": 0.6230239272117615, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/76d29a90032b3ec5b70d378a54fcb295040c9d42024136f939eada7f44c28ca9.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "1254a0aaeab79b2e6c7c281ed67a9801e1326db1e3d95d7a6e84127a9aad16ca", "generation_seconds_batch": 14.769946539076045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864943981170654, "target_z": -1.093749985401768, "ce": 6.173524379730225, "genuineness": 4.356721878051758, "identity": 0.7213891744613647, "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_RANT_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "77aebbbc51418aa0d11c9e322b972bf8b4b14212f8ecf6a374c7c7f8ca2cab73", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ec26381b5ab871ae8dc0148272a9d142b4b9fe3d3f39c3966847863b2a2ab176", "generation_seconds_batch": 17.44220868300181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.268369674682617, "target_z": 1.4374999941607072, "ce": 6.117925643920898, "genuineness": 4.093128204345703, "identity": 0.5674480199813843, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/77aebbbc51418aa0d11c9e322b972bf8b4b14212f8ecf6a374c7c7f8ca2cab73.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "7834340a75f877a0be447c237a8e0f7e4255dfcf1a8c6795fd7addbec2456efa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "61376e885ae32541b6b64e361ffddb85b9f3426256642886190f32c2a9800e55", "generation_seconds_batch": 26.91664115001913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8497023582458496, "target_z": 1.109375091968861, "ce": 6.157639026641846, "genuineness": 3.6192874908447266, "identity": 0.605567455291748, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7834340a75f877a0be447c237a8e0f7e4255dfcf1a8c6795fd7addbec2456efa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "7844de9901fb2251a395c7c5ee581f7fa0a9814f79c5aab2cdb242a3a50e8b5d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "cc630eeb8169a5de9fd4dd8b92c3f55692c87b13321cec8d105a8d4710ef00ec", "generation_seconds_batch": 20.192575825029053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2323499917984009, "target_z": -0.15820316459770403, "ce": 6.181467056274414, "genuineness": 4.394378662109375, "identity": 0.38728755712509155, "hypothesis": "J'ai trouvé la quille de boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7844de9901fb2251a395c7c5ee581f7fa0a9814f79c5aab2cdb242a3a50e8b5d.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": "7846abe5a5c102d84992b037e725b3c1f68e44637a776e3b58b3d138e20ce630", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f59c0e8f34d1a11f436b11310114582195d35f204235bca861ea0b10af301b26", "generation_seconds_batch": 18.32488360791467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1886234283447266, "target_z": 1.3749999416070724, "ce": 5.919358730316162, "genuineness": 3.7918784618377686, "identity": 0.6557384133338928, "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_RANT_reg/7846abe5a5c102d84992b037e725b3c1f68e44637a776e3b58b3d138e20ce630.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": "78c8c88b98c084a1f072bdc7b9f68fb44b0c453ac6ceda82f69e38a4717a7131", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc4977e5e87749857e1314e310c0077062700171036221b6b289c447b504b1e3", "generation_seconds_batch": 16.398936051991768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.7237372398376465, "target_z": 2.578124832120333, "ce": 5.661221981048584, "genuineness": 3.47493839263916, "identity": 0.6046063303947449, "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_RANT_reg/78c8c88b98c084a1f072bdc7b9f68fb44b0c453ac6ceda82f69e38a4717a7131.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "784bdb430e6f4d961aa6a912090b6d8836adf8dacfe73e2358289ac9ca8de506", "generation_seconds_batch": 18.24091650999617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35763418674468994, "target_z": -0.8437499620445971, "ce": 6.0305562019348145, "genuineness": 4.331617832183838, "identity": 0.6414502859115601, "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_RANT_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "79e449365a686490c0a3b19f1f9c9946241476b5dde3082342f4889c4cdc36ab", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "479e5538866a93c79eb3109ca8a940f9e82b3e11eaf6f4c79ebed828bd4f8e5a", "generation_seconds_batch": 21.218388702953234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0280004739761353, "target_z": -0.31835936240902496, "ce": 6.09409761428833, "genuineness": 4.243753433227539, "identity": 0.6355329751968384, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/79e449365a686490c0a3b19f1f9c9946241476b5dde3082342f4889c4cdc36ab.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "7a4575aceb0fb72325fd0983f37304629711d5d80546a41c0ee3f4754a2e4f42", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "310e8b0a2581fc78941c57e2bef5dd60d112b892dd8c68d1b890d71612d9eccb", "generation_seconds_batch": 22.950990261044353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0653815269470215, "target_z": -0.28906246277450864, "ce": 6.197352409362793, "genuineness": 3.9927117824554443, "identity": 0.6295263171195984, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7a4575aceb0fb72325fd0983f37304629711d5d80546a41c0ee3f4754a2e4f42.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": "7a960782fbdb4f7a13e2fb44fadbbb9da172f08a58552b6f174be0e950b54965", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58851a93f5a8c3dcc148c3b4ff425d7b711ff58fdee948b7df9b5c0950af6531", "generation_seconds_batch": 21.039392792037688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8796072006225586, "target_z": 1.132812611676474, "ce": 5.839932441711426, "genuineness": 3.6412534713745117, "identity": 0.49933841824531555, "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_RANT_reg/7a960782fbdb4f7a13e2fb44fadbbb9da172f08a58552b6f174be0e950b54965.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": "7aac05cc7354261962fef79d006ba1f3bc95c4ba647bfc67a7fc5a97f11ca54a", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82cff2dd12ae9d3c3b2a4762a3a4198b22476474948f6b144219fcea81976cbb", "generation_seconds_batch": 16.903757876949385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.567417621612549, "target_z": 1.6718748175221012, "ce": 5.792276382446289, "genuineness": 3.6412534713745117, "identity": 0.5245058536529541, "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_RANT_reg/7aac05cc7354261962fef79d006ba1f3bc95c4ba647bfc67a7fc5a97f11ca54a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "7b1965ece42aa55c09f0310d6310023dddbf1f45ff98d481aa9b460381d45c98", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "464692f853b0500726de3429b03b69e7cf7b139da5a88c57b143c0e6309ade36", "generation_seconds_batch": 20.192575825029053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.297824501991272, "target_z": -0.8906250014598232, "ce": 6.1020402908325195, "genuineness": 4.293961524963379, "identity": 0.8021605014801025, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7b1965ece42aa55c09f0310d6310023dddbf1f45ff98d481aa9b460381d45c98.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": "7b3feb1d2813efead101f2328e6c2d6abcaa03d78a64dcaa2535157c89765c7f", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7e8781dadab4311aabfb269740d5335e334b15cd08d44a852b90259ce6b3a815", "generation_seconds_batch": 16.295550673035905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6054797172546387, "target_z": 0.9179688678807226, "ce": 5.752562999725342, "genuineness": 3.528284788131714, "identity": 0.3343205153942108, "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_RANT_reg/7b3feb1d2813efead101f2328e6c2d6abcaa03d78a64dcaa2535157c89765c7f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "7bcad29a01eead6f87dc0b7ddef0ae5ea45f47939bc4ee5aa00b70b8ce86d4b3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "71e29b28811658bf6aa3ef90124d9fdd4bce7d6a99b200083aee5b8fb8b9c72c", "generation_seconds_batch": 18.874024434015155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3924654722213745, "target_z": -0.032714801187030115, "ce": 6.117925643920898, "genuineness": 3.613011360168457, "identity": 0.5335102081298828, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7bcad29a01eead6f87dc0b7ddef0ae5ea45f47939bc4ee5aa00b70b8ce86d4b3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "7d1a82d4e893feb2c8fe624ff10dc20b720c8abc2b02243d3060685ff426296a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "82f4955f8db346d3d2ef4724621d26fa56447e9c13bea889dbb39af99b049f96", "generation_seconds_batch": 25.190292931976728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7300829887390137, "target_z": 1.0156250131384088, "ce": 5.5539960861206055, "genuineness": 3.7291181087493896, "identity": 0.7093687653541565, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "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_RANT_reg/7d1a82d4e893feb2c8fe624ff10dc20b720c8abc2b02243d3060685ff426296a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "7d1c4d2a7a99520bc013711d2f0670bc0f2511d36933b4b1502ae65d73163657", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cab1a117523e188429138f8fd24b82dc9234bd326dab3c8be10147a2c4de131e", "generation_seconds_batch": 22.827915480011143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049067735671997, "target_z": 1.2656250364955797, "ce": 5.816103935241699, "genuineness": 3.5439748764038086, "identity": 0.5251131057739258, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7d1c4d2a7a99520bc013711d2f0670bc0f2511d36933b4b1502ae65d73163657.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "7daf373289c0b54063b0e89c027e3117f79a9159cef23f72e45b26dfbde19cc4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6c6e835c2ca5f788dbca154554f1749a8077510393135852378085e3ab75847d", "generation_seconds_batch": 19.67135965800844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.704906940460205, "genuineness": 3.314899444580078, "identity": 0.43586692214012146, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7daf373289c0b54063b0e89c027e3117f79a9159cef23f72e45b26dfbde19cc4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "7db3a73a89285b0afa6df50fd70d5fe941f7899a3d56e1368d0a1a93911ab8bf", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "640b3984883aff7f397d13c0804eac3cb57ccea8dfb327ab5ad940e843af57a7", "generation_seconds_batch": 17.377502955961972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9507463574409485, "target_z": -0.37890624160601666, "ce": 6.157639026641846, "genuineness": 4.174716949462891, "identity": 0.5541974306106567, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7db3a73a89285b0afa6df50fd70d5fe941f7899a3d56e1368d0a1a93911ab8bf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "7df9c938850e62f68f6c9d9ef48d310158bf5b7b32f0ba5260d4f8665efecc48", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8ecb8d2c1eb73ab7c97bbcad63b34e7b50baa3e04c122cd8f57f89f4db1a7600", "generation_seconds_batch": 20.454556047101505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0927941799163818, "target_z": -0.26757816313788085, "ce": 6.141753673553467, "genuineness": 3.823258876800537, "identity": 0.36584970355033875, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7df9c938850e62f68f6c9d9ef48d310158bf5b7b32f0ba5260d4f8665efecc48.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "7f3587975df01732c15a0ff84e64609e15ab2c84a6b91dfe7889a66e8d920445", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e57221a7b762bc03ff0c208cd560d143d0ec2d2c9461f7730a2bef1550e97bf", "generation_seconds_batch": 23.163964190986007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8678274154663086, "target_z": 0.339843755474337, "ce": 6.117925643920898, "genuineness": 3.7667744159698486, "identity": 0.4184763729572296, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7f3587975df01732c15a0ff84e64609e15ab2c84a6b91dfe7889a66e8d920445.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "7f3a4ff730aa40d7573072c7aec3b3913161be0f5c157a2c8f018844426d3837", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "06e7c780d7df73cf68b026afb1e22c159d51ec7c143de185fce30b6555417434", "generation_seconds_batch": 21.335711191990413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.221701145172119, "target_z": 0.6171875518237232, "ce": 5.839932441711426, "genuineness": 3.71028995513916, "identity": 0.3003268837928772, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7f3a4ff730aa40d7573072c7aec3b3913161be0f5c157a2c8f018844426d3837.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "8389bd99f7462d61a13c56bb2c92aac2da661edfe429f960f7915177393e750d", "generation_seconds_batch": 15.710559557075612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32772934436798096, "target_z": -0.8671874817522102, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.37758249044418335, "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_RANT_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "7f949360add77b4cc002be2776e8949b8780180415f45ae42e6c2f2ebfc13d6e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9f0d5333c2b15ceab101a2a43abb8a8ec9b5f68d5af842ffcc2d300e05c68b20", "generation_seconds_batch": 15.563130744034424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.990619421005249, "target_z": -0.34765626204354133, "ce": 6.157639026641846, "genuineness": 3.886019229888916, "identity": 0.3847702145576477, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7f949360add77b4cc002be2776e8949b8780180415f45ae42e6c2f2ebfc13d6e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "7f9db5989b28dfffde132014d4bf4c7a5c8471c0de8de8c491a0e29164190cd6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f2e2e83825cb1ddedfacfaab73c95538cc3554838b2e2429f412bab5500cfc85", "generation_seconds_batch": 21.96912500599865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.607290744781494, "target_z": 1.7031248437989186, "ce": 5.792276382446289, "genuineness": 3.4090399742126465, "identity": 0.6675699353218079, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/7f9db5989b28dfffde132014d4bf4c7a5c8471c0de8de8c491a0e29164190cd6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "8000f18072040804353f97144f65224c2dcf57ba7ee888eda7035de17566d751", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2e6b5211a4df0900f9bc167255d80991c5944b49884e63f7baca439df35120e5", "generation_seconds_batch": 16.85546148591675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0255084037780762, "target_z": -0.3203124890513261, "ce": 6.054384231567383, "genuineness": 4.256305694580078, "identity": 0.6858503818511963, "hypothesis": "La porte 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_RANT_reg/8000f18072040804353f97144f65224c2dcf57ba7ee888eda7035de17566d751.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "805c2d2d2e73fb14d8f97e8319e61896a79f1654401efc531f839e5ca73f6c36", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "783d51167b03f34070d1fb0e445af5902b438ed12b702d529f5c0cbe7778b017", "generation_seconds_batch": 16.36142500001006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7781128883361816, "target_z": 0.26953119635149775, "ce": 6.078212261199951, "genuineness": 3.8671910762786865, "identity": 0.6068508625030518, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/805c2d2d2e73fb14d8f97e8319e61896a79f1654401efc531f839e5ca73f6c36.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "8097b6fa41a2847b4edbb3c5807cfad4dad91fc5daa62ba0a325fc66b0d81438", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "495dda6f457c1430602ec3f6eac2a0e8848551ea2d5ef82eb064066214707b31", "generation_seconds_batch": 22.60545622196514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7069735527038574, "target_z": 1.7812499094909622, "ce": 5.542081832885742, "genuineness": 3.625563621520996, "identity": 0.4827538728713989, "hypothesis": "¡La puerta sigue abierta! ¡He dejado tu abrigo sobre la silla! Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8097b6fa41a2847b4edbb3c5807cfad4dad91fc5daa62ba0a325fc66b0d81438.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "3f0d60914eaec2da630d04b9e8bf88ac3146a2e5a9ef360a80afacfc833b8a5d", "generation_seconds_batch": 17.694810020038858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.1407794952392578, "target_z": -1.2343750102187623, "ce": 6.324435234069824, "genuineness": 4.193545341491699, "identity": 0.6305800080299377, "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_RANT_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "80cd93a9599866234696be557ed87192aa3b119c65f678088cca1233a9833b07", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cf35c5c5ea6a8a9f07de158f0ee475bdfed3672536c2523a85b22c26f9d631d6", "generation_seconds_batch": 21.9715138199972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4204232692718506, "target_z": -0.010803252177752557, "ce": 6.070269584655762, "genuineness": 4.093128204345703, "identity": 0.6315740346908569, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/80cd93a9599866234696be557ed87192aa3b119c65f678088cca1233a9833b07.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "817b2339e7dfded5a6fa18d332c6b061ef01644181e343842ad265020ac9e007", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8b04bf5a6b74c58bdb817d343d606623f09b4dd49f143532a1a681aaa20172a", "generation_seconds_batch": 21.88931135996245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.650336742401123, "target_z": 0.9531249605847739, "ce": 6.054384231567383, "genuineness": 4.049196243286133, "identity": 0.11681094765663147, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/817b2339e7dfded5a6fa18d332c6b061ef01644181e343842ad265020ac9e007.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "81855e8d6cfe9601989876da654ea5c3ef687d13807945a543c08e55c62749d5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "337dd49ab4f5bcc23257d02e9f9612f167250e44159c9619b11e2b4300559ef9", "generation_seconds_batch": 21.568647903972305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.271542549133301, "target_z": 0.656250084669745, "ce": 6.038498878479004, "genuineness": 3.860914945602417, "identity": 0.6391152739524841, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/81855e8d6cfe9601989876da654ea5c3ef687d13807945a543c08e55c62749d5.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": "81bfa10357bd54ce123b9b6913193dbdb951dc6a81fb54cd9ff5bf6c0a028700", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4edf913b810b837238c0b6ffd77fcea040aa52aaf4894294025f7a0941790f92", "generation_seconds_batch": 15.69900585594587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769956111907959, "target_z": 1.0468750394152262, "ce": 6.054384231567383, "genuineness": 3.8671910762786865, "identity": 0.5829533338546753, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. L'eau bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/81bfa10357bd54ce123b9b6913193dbdb951dc6a81fb54cd9ff5bf6c0a028700.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "81f4f93b3facf311cc1028c8e66303a84db417f1552a0cf9af55d41ce0cc644d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7358940e75d9d909bccc290008162c7a7c01e6e0fd363030da8aae290e7ab00b", "generation_seconds_batch": 19.692703776061535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344943046569824, "target_z": 2.281249956205304, "ce": 5.577824115753174, "genuineness": 3.6098732948303223, "identity": 0.28621354699134827, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt! Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/81f4f93b3facf311cc1028c8e66303a84db417f1552a0cf9af55d41ce0cc644d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "8227808fd395b8ea1b33402df5afb5d30ea270ed84b704a04fa13bb0280a1c28", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f26e8b7edf1a9f2e88e600cdc3fd4dd8ef5a3c19cbfa517641c27896cb6ca346", "generation_seconds_batch": 23.35153289604932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460939645767212, "target_z": 0.8046875226272594, "ce": 5.597680568695068, "genuineness": 3.8671910762786865, "identity": 0.7100796103477478, "hypothesis": "¡La puerta sigue abierta! He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8227808fd395b8ea1b33402df5afb5d30ea270ed84b704a04fa13bb0280a1c28.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "823c2cd6db1a56cc9b12d20cb0ab28ac240aaabd792848ced57248cde5e63f7b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "f3fca58d629c8a19e6ca7653d5809feaf547a0a3c9a3541c4ee303c6449083a6", "generation_seconds_batch": 20.373611327027902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.09093809127807617, "target_z": -1.1953124773727406, "ce": 6.117925643920898, "genuineness": 4.519899368286133, "identity": 0.5665073990821838, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/823c2cd6db1a56cc9b12d20cb0ab28ac240aaabd792848ced57248cde5e63f7b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "827afeb10bcecfe6dd2391ad518e6900514351b5aa20bf43395d0b1fedf7a0a8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88247027a392fdd653f73ae7eebd3aabb2cc48ddcb3eeafb1ada547f22c9e6f9", "generation_seconds_batch": 21.327272180002183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4876718521118164, "target_z": 1.6093751386832031, "ce": 5.700935363769531, "genuineness": 3.7542223930358887, "identity": 0.4816282093524933, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/827afeb10bcecfe6dd2391ad518e6900514351b5aa20bf43395d0b1fedf7a0a8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "82b21b7781a1bc34ea755a5471e2e4eb979e3fe6a9bfc07b6b814f594bde815f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "dc6936bc8a5ec6da2168d154ab670db375cc24e74038983c06f349ae75c9a0a4", "generation_seconds_batch": 22.741616470972076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.796238124370575, "target_z": -0.5, "ce": 6.165581703186035, "genuineness": 4.080576419830322, "identity": 0.42762935161590576, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/82b21b7781a1bc34ea755a5471e2e4eb979e3fe6a9bfc07b6b814f594bde815f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "82c186d84f956dae5d0745143e7819c55c353b75b3a58ef923a50be1d9ced8cf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f4c04a37ebe4f08d6f21cdf03a4585f1866bd48b42ef6bc2be1d66492f0332af", "generation_seconds_batch": 23.47590358403977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9756669998168945, "target_z": -0.3593750218973478, "ce": 6.070269584655762, "genuineness": 3.6914620399475098, "identity": 0.2516148090362549, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/82c186d84f956dae5d0745143e7819c55c353b75b3a58ef923a50be1d9ced8cf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "82e509c9553e39e0ebb34808d168cd9d07671d7b43c81f3fc835f78548223d23", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "49aa1d405134cf0671771cfa65a15fdee0bd1097105df961023fab5c2b3c0d0e", "generation_seconds_batch": 20.828897912986577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9557304978370667, "target_z": -0.37499998832141446, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.31049859523773193, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/82e509c9553e39e0ebb34808d168cd9d07671d7b43c81f3fc835f78548223d23.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "82fcf26f416af31c3859a91696d0664ab033cff289700c10b61b80d901c56005", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "cbee4bd59a3c874ecf4e0eccd6d843638b3d740b787bffb066ca4d446ddbe154", "generation_seconds_batch": 15.114476414979436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2821913957595825, "target_z": -0.11914063175168225, "ce": 6.173524379730225, "genuineness": 3.8797430992126465, "identity": 0.2625260651111603, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/82fcf26f416af31c3859a91696d0664ab033cff289700c10b61b80d901c56005.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": "831dc8e142c94e371e88d06f727df81f31ce227a467a21389f21b149e0e99a56", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c34f046e738e2a4b29d48cd0ed35acfd9fe40f4010f4c2be308cb1929bab0990", "generation_seconds_batch": 16.246230619959533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065831661224365, "target_z": 2.062500145982319, "ce": 5.967014789581299, "genuineness": 3.405902147293091, "identity": 0.5223273038864136, "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_RANT_reg/831dc8e142c94e371e88d06f727df81f31ce227a467a21389f21b149e0e99a56.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "833761e360a7e23df67302ee588bc7076bc6bda038031f3c667c615ff40750b4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4ea10cd4f7214855a4a98e814f7aff3b03a583bc360ac720bd8a24d02c2c77e7", "generation_seconds_batch": 22.98833900608588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431034803390503, "target_z": 0.7812500029196464, "ce": 5.974957466125488, "genuineness": 3.823258876800537, "identity": 0.6940304636955261, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/833761e360a7e23df67302ee588bc7076bc6bda038031f3c667c615ff40750b4.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": "8349a200e462000b40ee4ebaf1f88aa25cc76956d25d222a00a5d98bcb261f5f", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b819b457f547e26a5e920e8192f4c7a57cc95fdbfc548ef143b9993fadea330", "generation_seconds_batch": 30.06906650599558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0397801399230957, "target_z": 0.47460940036442795, "ce": 5.831989288330078, "genuineness": 3.961331605911255, "identity": 0.6665912866592407, "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_RANT_reg/8349a200e462000b40ee4ebaf1f88aa25cc76956d25d222a00a5d98bcb261f5f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "8384d965206813542d9267ff5a2bc0ac9e38bfc673754e5f8b716e7f2bfb18cb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e36b4227d993ea1baaea3b1ec76fcd61156b6694b89e9bcf3f89a0f0c6338c64", "generation_seconds_batch": 23.86017801403068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2067487239837646, "target_z": 0.6054687919699168, "ce": 5.808161735534668, "genuineness": 3.8420867919921875, "identity": 0.4790657162666321, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8384d965206813542d9267ff5a2bc0ac9e38bfc673754e5f8b716e7f2bfb18cb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "841a7e7c4e6ca844a4c3f694e9708d662533ed05a075d9bd71d28fe9b378b227", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "46c2962af4cdd2666cd0f15010bf3d93630d15a04abda3bb671aa7394a240423", "generation_seconds_batch": 26.610232162987813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0091946125030518, "target_z": 1.2343750102187623, "ce": 5.649308204650879, "genuineness": 3.5596652030944824, "identity": 0.4646766185760498, "hypothesis": "I found the little box! It was beside the old winto! We can look inside together!", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/841a7e7c4e6ca844a4c3f694e9708d662533ed05a075d9bd71d28fe9b378b227.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "842f517f7459e33c1826da90bfa2fdf1df5b6430431118af2a592f3f77c74d08", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4a1c7b3ec528514c5cdfe0cd39d33e5673422497341fefb622b4ff10c359dcca", "generation_seconds_batch": 16.449964698054828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3975071907043457, "target_z": -0.8125000291964638, "ce": 5.95112943649292, "genuineness": 3.427868366241455, "identity": 0.33248573541641235, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/842f517f7459e33c1826da90bfa2fdf1df5b6430431118af2a592f3f77c74d08.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": "85066a3247f8fdfa7269b2dec1e026c3e2342a0abc52fffd1b7b469b3c5150d4", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9aed5dd751a321cbf4d428605d72ecd406928efb101319ed7c82fc75f777c2f3", "generation_seconds_batch": 30.151650395942852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.241637706756592, "target_z": 0.632812564962132, "ce": 5.657250881195068, "genuineness": 3.7793264389038086, "identity": 0.7593486309051514, "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_RANT_reg/85066a3247f8fdfa7269b2dec1e026c3e2342a0abc52fffd1b7b469b3c5150d4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "851d1b67af612e1500120314c6d3cebb63be4f3a78fa7e357808a304e5df2ed5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "47e6836dc5cfef599cf984e77a0480ac40fe64d0e3dd686fb9e297624602d6a7", "generation_seconds_batch": 22.305283306981437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8759843111038208, "target_z": -0.43749999416070723, "ce": 6.0146708488464355, "genuineness": 4.2249250411987305, "identity": 0.35178300738334656, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/851d1b67af612e1500120314c6d3cebb63be4f3a78fa7e357808a304e5df2ed5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "854ccc96977f9e0a2c5324151d909156a3d401edaead67129ad69b6069068930", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78b8dd5565f802570a1e6a6221237d202e75ce456c49e6a036a1a998132ffccc", "generation_seconds_batch": 22.183919743052684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.511461615562439, "target_z": 0.060546832482649596, "ce": 5.5738525390625, "genuineness": 3.898571252822876, "identity": 0.5528901815414429, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/854ccc96977f9e0a2c5324151d909156a3d401edaead67129ad69b6069068930.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "85a7349510a9053fecd8948aeb60a725eed21a71b8f9123dc11dbf12e347479e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "820792697f73db58deea818e9b529511af95bb3a3d8e96c47bd4701d0f3b0a67", "generation_seconds_batch": 15.836484865052626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.16071605682373047, "target_z": -1.250000023357171, "ce": 6.1020402908325195, "genuineness": 4.582659721374512, "identity": 0.4220443665981293, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/85a7349510a9053fecd8948aeb60a725eed21a71b8f9123dc11dbf12e347479e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "862491a1bacfeef48ca185634e6a05b2bc58b05313f1c211fb72c6923f862575", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5e563f1cecd0fb10761851a7e96fc8849855347e104ecc4ffcca1d54912d5283", "generation_seconds_batch": 20.754544012947008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127002477645874, "target_z": 0.5429687394162819, "ce": 6.173524379730225, "genuineness": 4.099404335021973, "identity": 0.31207841634750366, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/862491a1bacfeef48ca185634e6a05b2bc58b05313f1c211fb72c6923f862575.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": "8642fbde7a36d593de78868269d1f2f3fe57b310f6258985ec6233caa6177afb", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b32459c690ffb7daa5cf9a8ecb2b3ba97510676a92f7f55353b4ddb19db589da", "generation_seconds_batch": 20.951573069090955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0291311740875244, "target_z": 1.250000023357171, "ce": 5.982900142669678, "genuineness": 3.7542223930358887, "identity": 0.6646413803100586, "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_RANT_reg/8642fbde7a36d593de78868269d1f2f3fe57b310f6258985ec6233caa6177afb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "86e21c233f7df8b796a5783848216fe05cc5a96e05ece6116d09fc65467fea50", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47e697d51d4e214ce2767af45ec0da022c3edff2f17325b3c72421167f9513cf", "generation_seconds_batch": 22.675156541983597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3532153367996216, "target_z": -0.06347656916044331, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.7430704832077026, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/86e21c233f7df8b796a5783848216fe05cc5a96e05ece6116d09fc65467fea50.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": "86eea2d63fb2b803d008198db08430babd617c1e89067b5dc3a626776b625f6d", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4994e310f7d97fd80debd3b90445488d7a2231fcc4f3cfa210de6ff78ee4201c", "generation_seconds_batch": 21.28093650494702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0390994548797607, "target_z": 1.2578125299263754, "ce": 5.7684478759765625, "genuineness": 3.722842216491699, "identity": 0.25516173243522644, "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_RANT_reg/86eea2d63fb2b803d008198db08430babd617c1e89067b5dc3a626776b625f6d.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": "86f627e68882bb157e26b71397fa9327ba22d0d2c1951b5a2dc31906532df2e4", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7550fb1bfaf6726ed5a02678315e8006ac2c89582450df27b005ae7d4b47dd52", "generation_seconds_batch": 23.016166558023542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3848161697387695, "target_z": 2.3124999824821217, "ce": 5.8875885009765625, "genuineness": 3.6192874908447266, "identity": 0.6035513877868652, "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_RANT_reg/86f627e68882bb157e26b71397fa9327ba22d0d2c1951b5a2dc31906532df2e4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "877a155796dfe2dae603ad3e1b641d72cbe070f886ccdc119966a742e54f90f3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9eab69009db1bda571d3c8ccfca8fff073234e5f89f035d3554c7f034616ad20", "generation_seconds_batch": 21.76518342399504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2484331130981445, "target_z": 1.4218749810222986, "ce": 5.426913261413574, "genuineness": 3.653805732727051, "identity": 0.3396289348602295, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/877a155796dfe2dae603ad3e1b641d72cbe070f886ccdc119966a742e54f90f3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "8792947dd69ec118a5eeb2f619533689a9c4eb93e2ba0d7fddda9df1ed46309d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c8865daf029f8b8ce603dee41362a5c0b1f452011d9f173599a1a319003f2b7", "generation_seconds_batch": 24.081040455959737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.667100429534912, "target_z": 1.7499998832141448, "ce": 6.070269584655762, "genuineness": 3.7918784618377686, "identity": 0.621850848197937, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8792947dd69ec118a5eeb2f619533689a9c4eb93e2ba0d7fddda9df1ed46309d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "87b5a9de1176e8e33d2131f206368c2a1475f591943330e43991ee32ca925fb8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ed0554ea03d5110b8b4b392a6323e5eed616ca9498666ee556eb3110628f1c1a", "generation_seconds_batch": 17.087409920990467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.620431900024414, "target_z": 0.9296874408771608, "ce": 6.0305562019348145, "genuineness": 3.6224255561828613, "identity": 0.3827822208404541, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/87b5a9de1176e8e33d2131f206368c2a1475f591943330e43991ee32ca925fb8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "87eecce46667da26a0d2b74f0594d81156e139e1f38d6c750707539d0e436ea6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fc2ae6aeb7d21f8d402a03b5a8ef77cc246bf5f5510a30783834795255ddc07", "generation_seconds_batch": 20.63676712499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1057047843933105, "target_z": 2.0937501722591363, "ce": 5.6413655281066895, "genuineness": 3.377659797668457, "identity": 0.43730705976486206, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/87eecce46667da26a0d2b74f0594d81156e139e1f38d6c750707539d0e436ea6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "88ac1045389918f38283eaf3e1502ce150d040e9dd0245102e2730828967dd8e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "98aad7aaeb079f7c3cbe5d6a42d85e77916e5596d8f1156c1283f35bd0aff6b2", "generation_seconds_batch": 20.404730860027485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13833212852478027, "target_z": -1.0156250131384088, "ce": 6.0146708488464355, "genuineness": 4.344170093536377, "identity": 0.6356226801872253, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/88ac1045389918f38283eaf3e1502ce150d040e9dd0245102e2730828967dd8e.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": "88bbbf6df0d0da5b465ab011ef9a78a5a44cff922e9cbc9d0758525962967a36", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b465aeae8e6b4efb9b06b41d3699aff5af35febb76a2a2be9b3cead2141912ad", "generation_seconds_batch": 21.62205734103918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4808762073516846, "target_z": 0.8203125357656682, "ce": 6.0305562019348145, "genuineness": 3.653805732727051, "identity": 0.45194223523139954, "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_RANT_reg/88bbbf6df0d0da5b465ab011ef9a78a5a44cff922e9cbc9d0758525962967a36.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "8907d14f5ac966e2cac4f8b8717567b51db3e140e85939e75053366dbb96d795", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "732bdfb1b3020f9c568eb760d5e9510379af4d3905f37c686cbf0d31f7702c6c", "generation_seconds_batch": 23.153050375985913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.500812530517578, "target_z": 0.8359373620467085, "ce": 5.974957466125488, "genuineness": 3.7165660858154297, "identity": 0.4788421392440796, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8907d14f5ac966e2cac4f8b8717567b51db3e140e85939e75053366dbb96d795.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "89378c993f971bb1f634179e3148e2c91b65d0583c11af2ede822521afeee4d6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4bddde313aa53e29399b50d11c0b70bc9ad93424cfb6b4bde187fd5be96ca10d", "generation_seconds_batch": 17.329495823010802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0390994548797607, "target_z": 1.2578125299263754, "ce": 5.764476776123047, "genuineness": 3.760498285293579, "identity": 0.5906537175178528, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/89378c993f971bb1f634179e3148e2c91b65d0583c11af2ede822521afeee4d6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "89f8227f968a95899bced0d6bea293eebb444cd137cd7eaedfccfcbf87c8147a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42c7334ddbf830d7a2a5d903583fe60993b8463f20af48b53aaa83b5c03c6631", "generation_seconds_batch": 16.21295859606471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2859294414520264, "target_z": -0.11621098850257269, "ce": 5.839932441711426, "genuineness": 4.080576419830322, "identity": 0.5870273113250732, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/89f8227f968a95899bced0d6bea293eebb444cd137cd7eaedfccfcbf87c8147a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "8a21643426e93455e7f15b436124023bc09d4ce84d952b2983b6c93521b4bd08", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1b86e206ea6fb24dca9d12ffb1f8154f5f5e23c692a2953997179ede8fb24d71", "generation_seconds_batch": 21.991229766979814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.102081775665283, "target_z": 0.523437472993271, "ce": 6.157639026641846, "genuineness": 3.9425034523010254, "identity": 0.5262500047683716, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8a21643426e93455e7f15b436124023bc09d4ce84d952b2983b6c93521b4bd08.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": "8a51ce0094a02ae3cdd8d97fc6843bc3300d6ddf7ee5224d40a529a7298020a1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a851969fa6df04f06f3c9c588e74ac5a66714a8baf852a86ad3f62bb7f2a3865", "generation_seconds_batch": 20.714295544079505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 5.816103935241699, "genuineness": 3.8546390533447266, "identity": 0.32834962010383606, "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_RANT_reg/8a51ce0094a02ae3cdd8d97fc6843bc3300d6ddf7ee5224d40a529a7298020a1.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": "8b7449054f7dcd6d0dd0dab7a6a9fda3bf49885c646ca2a955b1159f0683e698", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cace4020a62ad4af7866cb35b9e8acac48637ae754a3715d5a16fcd8213f4aca", "generation_seconds_batch": 24.592902717995457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7468466758728027, "target_z": 1.8124999357677796, "ce": 6.006728172302246, "genuineness": 3.613011360168457, "identity": 0.5846618413925171, "hypothesis": "Je bericht kwam dat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8b7449054f7dcd6d0dd0dab7a6a9fda3bf49885c646ca2a955b1159f0683e698.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "8c22a33a1291e9476c4d8484dcc5f5f9ee70814736ccb45a0b7f5e1de267546d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4589f25f618365eb255186045ec30296b9fef7c2c372f77486db30bcd6cacf30", "generation_seconds_batch": 22.6374180810526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.19179630279541, "target_z": 0.5937500321161102, "ce": 5.5817952156066895, "genuineness": 3.7416701316833496, "identity": 0.478421151638031, "hypothesis": "I found the little box. It was beside the old window. We can look inside together!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8c22a33a1291e9476c4d8484dcc5f5f9ee70814736ccb45a0b7f5e1de267546d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "8c4b2d3deb31f766003b57c9996a3ba35aca59b2dba0fffa77fdf62a3318d66d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "51b1b9b75b0153957388331b73a1d7e4c5523c8398b33fe99cf6d679a95aa6ff", "generation_seconds_batch": 18.43499487999361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1487503051757812, "target_z": 1.343749915330255, "ce": 5.967014789581299, "genuineness": 3.5690791606903076, "identity": 0.5167568325996399, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8c4b2d3deb31f766003b57c9996a3ba35aca59b2dba0fffa77fdf62a3318d66d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "8c59f0da26c6f7a38a99e9d6761e22ee8f485e48482fe87fa77dafc25fa49662", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6f9e6e17ae9ebd6a42d77a003ce3f3ab816d0a479a1f571da449b82ebde56b3d", "generation_seconds_batch": 23.51941510697361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7314443588256836, "target_z": -0.5507812459854863, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.5002841949462891, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8c59f0da26c6f7a38a99e9d6761e22ee8f485e48482fe87fa77dafc25fa49662.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "8c80add65b6f29c0a5177e551616c5d2167421e517e1174c454fc78595ba02c3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8de8b144ad2bfbbaebe1665b1c3ca340d8b6d170d980564dee55b1d62739a5ef", "generation_seconds_batch": 25.59376314491965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.108877182006836, "target_z": 1.3124998890534376, "ce": 5.77639102935791, "genuineness": 3.8169827461242676, "identity": 0.500420331954956, "hypothesis": "J'ai trouvé la petite boîte. Elle était prêt à la vie fenêtre. We've regarded ensemble what it contiens.", "wer": 0.47368421052631576, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8c80add65b6f29c0a5177e551616c5d2167421e517e1174c454fc78595ba02c3.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": "8d54f64bcfcf30c766beba2f47d4e2a50b795a7a704b63113ae1214537a1d0b2", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e6a1589380092b97111cdcf6b09f8a70b0c1194545cbc02b637c61c892b6dc2", "generation_seconds_batch": 21.73623162601143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5873541831970215, "target_z": 1.68749983066051, "ce": 6.062326908111572, "genuineness": 3.9425034523010254, "identity": 0.3555476665496826, "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_RANT_reg/8d54f64bcfcf30c766beba2f47d4e2a50b795a7a704b63113ae1214537a1d0b2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "8dee0a0fb06a98b37af524365806795481a03fbdf72bc50c825dd7d32aa377a2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "30ef7020d6c580b7caeec570214352033cc7665b5b9290fdecd2e6ea31a86153", "generation_seconds_batch": 14.752427338040434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1077466011047363, "target_z": -0.2558594032840743, "ce": 6.24500846862793, "genuineness": 4.143336772918701, "identity": 0.6413432359695435, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8dee0a0fb06a98b37af524365806795481a03fbdf72bc50c825dd7d32aa377a2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "8e0ee65fc9a90b3a2cb294a2401687c23093e99a8798ffda3c08962910bc97a4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "15538742673747c6caf2b87f1e88ec86dd78bcca354d95d41a10bed2de0aff94", "generation_seconds_batch": 26.864335351041518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.507608413696289, "target_z": 1.6250001518216117, "ce": 5.824047088623047, "genuineness": 3.7667744159698486, "identity": 0.4923478960990906, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8e0ee65fc9a90b3a2cb294a2401687c23093e99a8798ffda3c08962910bc97a4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "8e13fe5be90994afb28b647df3a7698a0f28af4099e7fad6a633c706499a3c30", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c78c215e58a5cc8bbd052026e5eeeb3c19f90c6a30604e1522494cf08de90f26", "generation_seconds_batch": 17.666048213024624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.919480323791504, "target_z": 1.1640626379532915, "ce": 5.919358730316162, "genuineness": 4.005263805389404, "identity": 0.5972121953964233, "hypothesis": "La porte 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_RANT_reg/8e13fe5be90994afb28b647df3a7698a0f28af4099e7fad6a633c706499a3c30.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "8e5ed7be535ed4a867147a27ef32d746122d22cc15018cd9b4efbca1480930ed", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "576f20556af4ddad3755decb729f30f1e5bfa19728350d1ade12d713541d295d", "generation_seconds_batch": 16.640784753020853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4976401329040527, "target_z": 1.6171876452524074, "ce": 5.7247633934021, "genuineness": 3.591045379638672, "identity": 0.6585882902145386, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8e5ed7be535ed4a867147a27ef32d746122d22cc15018cd9b4efbca1480930ed.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": "8edb840bfad18904f42e1152450da0688b2f859fb2a4bbfb1c924aaa6b074fe4", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00e160f6e8dc501bc71c758d784a8e3acc81f99766e30945d70a7cac5c5d526a", "generation_seconds_batch": 23.016166558023542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.138782024383545, "target_z": 1.3359374087610507, "ce": 6.173524379730225, "genuineness": 3.9550557136535645, "identity": 0.3280786871910095, "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_RANT_reg/8edb840bfad18904f42e1152450da0688b2f859fb2a4bbfb1c924aaa6b074fe4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "8eea6784d91867455bd6b2dc4f90dcdc47b8fda53b929b4a03df87d699dd4cd8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d724732196cc9078e87a9a492550b0e74c70d5de12938757c5b642fda03c65f8", "generation_seconds_batch": 15.841209507081658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.667100429534912, "target_z": 1.7499998832141448, "ce": 5.538110733032227, "genuineness": 3.4435582160949707, "identity": 0.3429337739944458, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8eea6784d91867455bd6b2dc4f90dcdc47b8fda53b929b4a03df87d699dd4cd8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "8ef3ca7d014a541d14b3e362b8c0306df2421ddf8f85c6ed825872b416eaedcf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "cb894b49199455ed2ba60d5cff7dca1a1de2578d12228635cb323fb58723a991", "generation_seconds_batch": 14.957987376954406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8361111879348755, "target_z": -0.4687500204375247, "ce": 5.990842819213867, "genuineness": 4.105680465698242, "identity": 0.43221092224121094, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8ef3ca7d014a541d14b3e362b8c0306df2421ddf8f85c6ed825872b416eaedcf.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": "8ef6262e1e352720e9bfd67314f3eb224f575213c5b488de6719508f08e9848c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69f60ef1a29e241245602f737fbe9b9df76ca6221ae87fa4a496b3da6e33688f", "generation_seconds_batch": 20.998677960014902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8397340774536133, "target_z": 1.1015625853996567, "ce": 5.681078910827637, "genuineness": 3.6977379322052, "identity": 0.6260984539985657, "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_RANT_reg/8ef6262e1e352720e9bfd67314f3eb224f575213c5b488de6719508f08e9848c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "8efc996f03d37260a6dffa54e9635ebbcacd27f4e53ed65763d51f57a010ffeb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "37312753659c42d821a636786ec529f855edda0ddec405d06e9631c69fa09309", "generation_seconds_batch": 23.13543946109712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6647238731384277, "target_z": 0.18066407426982448, "ce": 5.998785495758057, "genuineness": 3.7291181087493896, "identity": 0.5455543398857117, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/8efc996f03d37260a6dffa54e9635ebbcacd27f4e53ed65763d51f57a010ffeb.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": "90a9db137f68a940cf37c8f1c578600be2bc8c9dcd25390fbaa71380ce873b04", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fb18be5cc8a110a16a0585992ff0d7151da547aaef8b7596e8f8be1137e4d0f", "generation_seconds_batch": 21.8379767970182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 6.109982967376709, "genuineness": 3.613011360168457, "identity": 0.4102708697319031, "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_RANT_reg/90a9db137f68a940cf37c8f1c578600be2bc8c9dcd25390fbaa71380ce873b04.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "9184bac15e9d102735ab75280d0b0bc5655a0b9f21e68b11c68e08a94d903671", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ef15c81cf269983849cc4f75e5f9d0220d8a88bf47df736932d2e68b1598e5f3", "generation_seconds_batch": 25.156674116966315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1587185859680176, "target_z": 1.3515624218994593, "ce": 5.7128496170043945, "genuineness": 3.6412534713745117, "identity": 0.7148290276527405, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9184bac15e9d102735ab75280d0b0bc5655a0b9f21e68b11c68e08a94d903671.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "919757bf7e88c69f68d9e282c95d8aad20a3ec0a8643323627494ebe415396e3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0e44d5495b5b45e9585a4f1ef293363cbad83e47af6e67c0310eb87535080c21", "generation_seconds_batch": 25.31258234498091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3476659059524536, "target_z": -0.8515624686138015, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.5564312934875488, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/919757bf7e88c69f68d9e282c95d8aad20a3ec0a8643323627494ebe415396e3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "9220a0f00f8e14ef6918e14cd0b22508b0386151017cd7636f26e1c590868320", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a9cb7a8d6ecfc9d5452b5adfc32493b54dc14ef2ec7fbc3084aa5ab0ce5aef8", "generation_seconds_batch": 16.21295859606471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6118251085281372, "target_z": -0.6445312313872543, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.44863831996917725, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9220a0f00f8e14ef6918e14cd0b22508b0386151017cd7636f26e1c590868320.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "923d6f0bac10572f578ca0b5739e49f0e8067ee44a014a6ec45ceaf197821ee8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6715125f39341fe2c9d33cc0626867bf710e692ad1da7ba602eebf09911771a", "generation_seconds_batch": 21.76518342399504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0989089012145996, "target_z": 1.304687382484233, "ce": 5.462655067443848, "genuineness": 3.653805732727051, "identity": 0.5990545749664307, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/923d6f0bac10572f578ca0b5739e49f0e8067ee44a014a6ec45ceaf197821ee8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "924e279a03092f5f26e01906ef875b16aae315f1bc6b16092aa38dbca3923d5b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "862bfcd824cc8a919889b584017ac4e6297b836d90d2a88702bd901b22772b59", "generation_seconds_batch": 17.660610841936432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3082427978515625, "target_z": 1.4687500204375246, "ce": 5.804190158843994, "genuineness": 3.625563621520996, "identity": 0.5636938810348511, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/924e279a03092f5f26e01906ef875b16aae315f1bc6b16092aa38dbca3923d5b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "92807f44598cb2a371395bb5a0a3072520ee2fb4446f7bf6c6549e865e8561e7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30275fdb24c599fc182a5e52e7cfa4d6a7d8b9b4ec3be510cd172462b98ea95f", "generation_seconds_batch": 18.307726435014047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769956111907959, "target_z": 1.0468750394152262, "ce": 5.7684478759765625, "genuineness": 3.7416701316833496, "identity": 0.7587137222290039, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/92807f44598cb2a371395bb5a0a3072520ee2fb4446f7bf6c6549e865e8561e7.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": "929b19b3290534644b02351cbb6ca86443c9e720fe26d90f4476cf99c5824cda", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d64a11da0d7ccddf74b1c26d90bc807faf9922c99dc2159cc6945ce3e29b33c", "generation_seconds_batch": 16.724314004997723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1057047843933105, "target_z": 2.0937501722591363, "ce": 5.863759994506836, "genuineness": 3.8420867919921875, "identity": 0.29654163122177124, "hypothesis": "Deine Nachricht kam, als ich schon weg war! Ich las sie zweimal, dann legte ich das Handy weg. Sagt mir, wann wir reden können!", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/929b19b3290534644b02351cbb6ca86443c9e720fe26d90f4476cf99c5824cda.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "92e6eb7ea1ebf2bbad919192ecf6c62d1d14307e672ea6c5614b7b63e8284714", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8f98af6c009eb426a647ea2b768a9618e849c4c3e060ba60c37594b2bd7fb8d7", "generation_seconds_batch": 21.884859057026915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1668756008148193, "target_z": 0.5742187656930993, "ce": 6.109982967376709, "genuineness": 3.785602569580078, "identity": 0.6509990692138672, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/92e6eb7ea1ebf2bbad919192ecf6c62d1d14307e672ea6c5614b7b63e8284714.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "934e007863373e3bd508f327688b11b1e33e2b60a70342aa693eff107bb4341b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b8bfaac9af18eec28df43e899634ffbba6f079605931453092679edf8d2e4f1", "generation_seconds_batch": 20.831367084058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.871703147888184, "genuineness": 3.393350124359131, "identity": 0.6454173922538757, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/934e007863373e3bd508f327688b11b1e33e2b60a70342aa693eff107bb4341b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "937b4beea850803e80209736efd0af6ee8f75e3401d34c9c3bb1b6117a3fa472", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1994f6564902354f27e0de42c3b54d57bf7bdaf7ae3e5bd4db743d2ad2c4d5e3", "generation_seconds_batch": 18.633861989015713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1288137435913086, "target_z": 1.3281249021918462, "ce": 5.7684478759765625, "genuineness": 3.7667744159698486, "identity": 0.6054174304008484, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/937b4beea850803e80209736efd0af6ee8f75e3401d34c9c3bb1b6117a3fa472.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": "9418e7e2dd6572ea09baae97101fefdcf3b2e8ec8047d6ba9d7767adde4d2c61", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e41f3d552a5551f37624779a6a1af75013d80e37b8af29680813c65ce5e68e4", "generation_seconds_batch": 16.903757876949385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.607290744781494, "target_z": 1.7031248437989186, "ce": 5.756534099578857, "genuineness": 3.6789097785949707, "identity": 0.086382195353508, "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_RANT_reg/9418e7e2dd6572ea09baae97101fefdcf3b2e8ec8047d6ba9d7767adde4d2c61.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "9441edc805a8c5891c7d5b49650bfcabe95c5e51610f2b70384977af0e0a7cdd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "380b497940fcebe0a66330ed2030d184bdf0cc7e48f19815694cf2fff7858cac", "generation_seconds_batch": 22.109787208028138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6933826208114624, "target_z": 0.20312498394194492, "ce": 6.046441555023193, "genuineness": 4.080576419830322, "identity": 0.6980960965156555, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9441edc805a8c5891c7d5b49650bfcabe95c5e51610f2b70384977af0e0a7cdd.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "94571f72b5b31ef08d76c44e75efbc5d4ea141caa145a70046e3be0e1f1f29ef", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e2df7df0953dbbf6ba613b8b0c1a9b6a7dccbeca07091a7546405d15cf84e54e", "generation_seconds_batch": 16.29250151803717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.150905609130859, "genuineness": 3.6475296020507812, "identity": 0.6158958077430725, "hypothesis": "¡La puerta sigue abierta! He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/94571f72b5b31ef08d76c44e75efbc5d4ea141caa145a70046e3be0e1f1f29ef.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": "945efb1fe89ef60c46e33b7ab33f9adfa57f95c2293fcb09673e686bff9a2b4f", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6576daccd73681e693c36075370594c78ca0ea061835d3840d8f0e2e89ba273b", "generation_seconds_batch": 20.729951173067093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7183033227920532, "target_z": 0.2226562503649558, "ce": 5.720792293548584, "genuineness": 3.8797430992126465, "identity": 0.7228202819824219, "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_RANT_reg/945efb1fe89ef60c46e33b7ab33f9adfa57f95c2293fcb09673e686bff9a2b4f.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": "94dde3df89f0cc1387cb6f61597c08547467e62b43cb6cf4fa3101144ab286e7", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f4002558f8f5add03c3d76fb0de9935009d0951fb763ae45e9b95b5c728ba5c", "generation_seconds_batch": 18.32488360791467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8696389198303223, "target_z": 1.1250001051072698, "ce": 5.748591423034668, "genuineness": 3.8922953605651855, "identity": 0.6806192398071289, "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_RANT_reg/94dde3df89f0cc1387cb6f61597c08547467e62b43cb6cf4fa3101144ab286e7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "94e5f5d75073e63b05b4472e52b23699cda99fc04ea72c7837b85cb88e7f5fdf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba8760952e57da64948ec01663288aa86d2eca66793652d85ba2b92e2e34a6a2", "generation_seconds_batch": 21.77322254993487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6666505932807922, "target_z": -0.6015624919709724, "ce": 5.959072113037109, "genuineness": 4.193545341491699, "identity": 0.7025229930877686, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/94e5f5d75073e63b05b4472e52b23699cda99fc04ea72c7837b85cb88e7f5fdf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "94f2586a33d9cdabc5d90fae4f627d001bdaf536aa508911a108b0130968c299", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "35bf978a2e43c4536b5c2dc7dc721580f9ef06b3f0e3242c2a25140182651d2d", "generation_seconds_batch": 22.6374180810526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4210665225982666, "target_z": 0.773437496350442, "ce": 5.967014789581299, "genuineness": 3.6663577556610107, "identity": 0.5101855397224426, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/94f2586a33d9cdabc5d90fae4f627d001bdaf536aa508911a108b0130968c299.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "954cf7457821f647084b48b25a5bd2ac3bf2f495d1af53822155d0ac4d72ca63", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "369b6f39fad100685278ca93d61c01cfa9b233a33265b436e93051ad8f382ca8", "generation_seconds_batch": 25.651038215029985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.208559989929199, "target_z": 1.3906249547454812, "ce": 5.804190158843994, "genuineness": 3.478076457977295, "identity": 0.5777890086174011, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/954cf7457821f647084b48b25a5bd2ac3bf2f495d1af53822155d0ac4d72ca63.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "95632adcd5cc7064ea3a1286b70228ecb3bdec6b0369c9d5518cf070b4000fe0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "939a34fb09a70dab90526a4d49ca17418005613ae0afef90b04a65ca0da1d4c1", "generation_seconds_batch": 20.74181874305941, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7314443588256836, "target_z": -0.5507812459854863, "ce": 5.919358730316162, "genuineness": 4.017816066741943, "identity": 0.710726261138916, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/95632adcd5cc7064ea3a1286b70228ecb3bdec6b0369c9d5518cf070b4000fe0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "956cad767b80787ce8b69d827d475b37ad203729b170af72a632ce0d9fa5061e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "89ada8c8cf8cf5783c9bee7a4f92df68527d40e24f03071785f2f3c8bb651bdf", "generation_seconds_batch": 21.530670220032334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.085768222808838, "target_z": 2.0781251591207277, "ce": 6.062326908111572, "genuineness": 3.5565271377563477, "identity": 0.6508521437644958, "hypothesis": "La porte 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_RANT_reg/956cad767b80787ce8b69d827d475b37ad203729b170af72a632ce0d9fa5061e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "95a4da7a047ee5ff6842d41e2bab07daf98daf215d62ae311bcc0a66edba11af", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d72dbfdd1781637ec8be384098568045a85df449c3da710e3bbed70aa92e204d", "generation_seconds_batch": 20.836596430046484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.251605987548828, "target_z": 0.6406250715313363, "ce": 5.998785495758057, "genuineness": 3.9299514293670654, "identity": 0.5867383480072021, "hypothesis": "J'ai trouvé la 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_RANT_reg/95a4da7a047ee5ff6842d41e2bab07daf98daf215d62ae311bcc0a66edba11af.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "95c6a836fe9c2b8c0cd82886bcc8ebd24f77b15c9ae08cb51a887f2c5f5f6f1f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5d4d04e7d450278c5f7f79ca5e1d7cf69318ffa754eed564ff024b6aa7b9c0dd", "generation_seconds_batch": 25.13324150198605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4677352905273438, "target_z": 1.5937501255447943, "ce": 5.665193557739258, "genuineness": 3.4121780395507812, "identity": 0.46383053064346313, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/95c6a836fe9c2b8c0cd82886bcc8ebd24f77b15c9ae08cb51a887f2c5f5f6f1f.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": "95eb03680d31865e4bc495314bf7a49113e30a60d1de6a03e7f32e4d6e99667c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99f200253ed3e7beb8178611efc6feea9b3271f7325412237a00c50463d25efd", "generation_seconds_batch": 16.31859594595153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.447798728942871, "target_z": 1.5781251124063855, "ce": 5.919358730316162, "genuineness": 3.490628719329834, "identity": 0.3601342439651489, "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_RANT_reg/95eb03680d31865e4bc495314bf7a49113e30a60d1de6a03e7f32e4d6e99667c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "95f08363eceac4d89476ca6b34ba2e09714e9b2b228debf57862d2651e1e335c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7826f3abc4edcc08f0101474ed9297500f0afd27fcec0c69f96709f7da082785", "generation_seconds_batch": 20.11865293898154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120500564575195, "target_z": 0.5312499795624753, "ce": 5.95112943649292, "genuineness": 3.873466968536377, "identity": 0.6036390066146851, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/95f08363eceac4d89476ca6b34ba2e09714e9b2b228debf57862d2651e1e335c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "96538e9ce6788ba9432eaf22a4a23396015cfc799edc4cff0bc88cbd3cda4f0a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "94fe35582414b9bdeaa8b9f8474b3db07251843e1d3f32b3bef2bb61d2df8309", "generation_seconds_batch": 22.109787208028138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431034803390503, "target_z": 0.7812500029196464, "ce": 5.959072113037109, "genuineness": 3.7165660858154297, "identity": 0.6930339336395264, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/96538e9ce6788ba9432eaf22a4a23396015cfc799edc4cff0bc88cbd3cda4f0a.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": "9660b63518489190e7c1323e9ad559cf84afc45c9cc271ada35b9ae30c3fd5ec", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de249a1acdd80d842550a7981449f8326b43ee75204e7ccab55261eb0bf4f874", "generation_seconds_batch": 30.235761043033563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1070659160614014, "target_z": 0.5273437262778732, "ce": 5.974957466125488, "genuineness": 3.6789097785949707, "identity": 0.5658695101737976, "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_RANT_reg/9660b63518489190e7c1323e9ad559cf84afc45c9cc271ada35b9ae30c3fd5ec.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "9663e5753bd1c8c312c778d4e767eed2e93272cf61f7e22d269b0c41accee5a5", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1c2bc9ee09746f4261dfda4d38df0cf858a1eb956ab29dcfeb1d2d1beed529fe", "generation_seconds_batch": 18.07022723800037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5769361257553101, "target_z": -0.6718750043794696, "ce": 6.181467056274414, "genuineness": 3.7667744159698486, "identity": 0.2964960038661957, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9663e5753bd1c8c312c778d4e767eed2e93272cf61f7e22d269b0c41accee5a5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "96fb37a493a892132f550640976e4f2f60f421562cb64a89d251c2047558ebb7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "15b3b556d23bf03dcf7288ca49f5f140eb07e0cae20be62e1f9447e62f5e26cc", "generation_seconds_batch": 18.447684631915763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.530717372894287, "target_z": 0.8593748817543216, "ce": 5.959072113037109, "genuineness": 3.6412534713745117, "identity": 0.5605491399765015, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/96fb37a493a892132f550640976e4f2f60f421562cb64a89d251c2047558ebb7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "97a1467f74f03787add33632a4f7f3cc7b48589849ed39954af616c0242654e9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "788a151a49784e8cf6483cddc87d1bfec1f813cd6f1b44e02290507212e6d1a5", "generation_seconds_batch": 16.025872732978314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8098292350769043, "target_z": 1.0781250656920436, "ce": 6.038498878479004, "genuineness": 4.055472373962402, "identity": 0.6625797748565674, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/97a1467f74f03787add33632a4f7f3cc7b48589849ed39954af616c0242654e9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "97c62ab49e06b1d3dc389601fb3780ee0d2b1ddee5338ad83ad08c0f0cb8ec14", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7289d79fe22f98bd20d038f91bd71eb35619ea08890f5ce722a2b4f5894ac517", "generation_seconds_batch": 21.04585503297858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6865871548652649, "target_z": -0.5859374788325638, "ce": 5.824047088623047, "genuineness": 4.369274139404297, "identity": 0.5441874861717224, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/97c62ab49e06b1d3dc389601fb3780ee0d2b1ddee5338ad83ad08c0f0cb8ec14.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "97dccff53a09cef42a467c0718b1294944c75bcf1db2af305d734ec9a9bd0dfe", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "3397bf189912d25803f50fe682bffaab5e96f0db79f5cc42dfbc84d6e8b72967", "generation_seconds_batch": 24.284205751959234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8197975158691406, "target_z": 1.0859375722612479, "ce": 5.71682071685791, "genuineness": 3.515732765197754, "identity": 0.5840011835098267, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/97dccff53a09cef42a467c0718b1294944c75bcf1db2af305d734ec9a9bd0dfe.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "982c3fb1503501fd6be489fee6fc6d92ec1e8d9c038b4836333c222f05cda637", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46dfa368f738a5d5085632b49314d021ca7dca0ea77a2fbc6e7ca0134c0e505d", "generation_seconds_batch": 23.764851261978038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3974494934082031, "target_z": -0.0288086413311121, "ce": 6.133810997009277, "genuineness": 3.9550557136535645, "identity": 0.4524904787540436, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/982c3fb1503501fd6be489fee6fc6d92ec1e8d9c038b4836333c222f05cda637.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "98d2a4073baaaddc1292ca07946ff54c8c6cc3dd79e1893c39dd0e76a46a5e91", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bc01f3ae74da0859e6d3da32e6567ccb935f2720db38b8da25ec6c1af5314b19", "generation_seconds_batch": 20.754544012947008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7631605863571167, "target_z": 0.2578125299263754, "ce": 6.284721851348877, "genuineness": 4.331617832183838, "identity": 0.3330441415309906, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/98d2a4073baaaddc1292ca07946ff54c8c6cc3dd79e1893c39dd0e76a46a5e91.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "995b9ee1c7260b53e2590924d79d448787913a4b19833e5d33de84633890d9be", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "978ccd1dcf2dc1fac80d148bda4a439edff93dd81d0500a70fd0f0e2b0d9de7b", "generation_seconds_batch": 20.937381447991356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.839932441711426, "genuineness": 3.3211755752563477, "identity": 0.4065871238708496, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/995b9ee1c7260b53e2590924d79d448787913a4b19833e5d33de84633890d9be.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "99ae4be8c377bd58962c538980850ad96ee3f2a83e2fd499ad07881b330332c6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "56b55dc9ec5de109e14f06ae0b19c936e012641808dd19884a0fbb462040184f", "generation_seconds_batch": 19.130376418004744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.407925605773926, "target_z": 1.5468750861295681, "ce": 5.9431867599487305, "genuineness": 3.8169827461242676, "identity": 0.467379093170166, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/99ae4be8c377bd58962c538980850ad96ee3f2a83e2fd499ad07881b330332c6.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "c420cdacf229c16182217420912337f7ab84556bc63e0246efbf619f03aa7e50", "generation_seconds_batch": 15.23482078302186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32772934436798096, "target_z": -0.8671874817522102, "ce": 6.340320587158203, "genuineness": 4.545003414154053, "identity": 0.6758453845977783, "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_RANT_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "9a1740f88175c25f470104c062a8b5f7821a5cc9cabec5067e34fe8085df8ff4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17eebaf493be63ea282f1d3866de6d0dea8dc4ef72d1be7886bf5643a7d48472", "generation_seconds_batch": 22.41062505193986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.267238974571228, "target_z": -0.1308593916054888, "ce": 6.229123115539551, "genuineness": 4.20609712600708, "identity": 0.7065802216529846, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9a1740f88175c25f470104c062a8b5f7821a5cc9cabec5067e34fe8085df8ff4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "9a57c364f6c636307029e79f54a24da6ebcd23005972c6bf4757923f592a95bf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ed1dd7a378c64d15ed789695f7c8824bf650aaab1c7d2532bf9917692c8e829e", "generation_seconds_batch": 22.071215928997844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0454449653625488, "target_z": -0.30468747591291734, "ce": 6.2767791748046875, "genuineness": 4.18726921081543, "identity": 0.7309727668762207, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9a57c364f6c636307029e79f54a24da6ebcd23005972c6bf4757923f592a95bf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "9a6399e42a98ba0d88fb89661cc7c2e1e7515b82c04a8a82a70fd03edc9d12d0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "6ac53985219454923664062d0fdd07be9b4e2911c8f255a032ab8fd4b9569871", "generation_seconds_batch": 25.183253731927834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3213837146759033, "target_z": 0.6953124306583984, "ce": 5.700935363769531, "genuineness": 3.7040140628814697, "identity": 0.4881701171398163, "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_RANT_reg/9a6399e42a98ba0d88fb89661cc7c2e1e7515b82c04a8a82a70fd03edc9d12d0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "9af076551a09599b36f091c2e48c5d7ae6d7b0751a43e6f53bb7b52534cbd41d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "25782643328f66e1b81820cde719a0917b179f6d44f41d68b1bd18b4bc2308ae", "generation_seconds_batch": 24.33940891502425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0690040588378906, "target_z": 1.2812498627766202, "ce": 5.752562999725342, "genuineness": 3.6789097785949707, "identity": 0.1563829779624939, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9af076551a09599b36f091c2e48c5d7ae6d7b0751a43e6f53bb7b52534cbd41d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "9b0a039a0c9aa3098835bed47ce2b1330e41cfb9a2e1247a01849d7c7104c9cb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "805120639d17d1ba16734809866a55ce3c3641c67a4a8eb6724a25b0f64246f8", "generation_seconds_batch": 24.045199324958958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0454449653625488, "target_z": -0.30468747591291734, "ce": 6.205295085906982, "genuineness": 3.8922953605651855, "identity": 0.5625258088111877, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9b0a039a0c9aa3098835bed47ce2b1330e41cfb9a2e1247a01849d7c7104c9cb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "9b142d4e95b00d4c2c01343b7488b975115b104f2c0688bcce3dc0121fdd1187", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5f0f5a0151b13d1552b152df4971d23705672b1adf40238d27c512504ab63585", "generation_seconds_batch": 23.78331406810321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.05473256111145, "target_z": 0.48632816021823444, "ce": 5.998785495758057, "genuineness": 3.7793264389038086, "identity": 0.6855337619781494, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9b142d4e95b00d4c2c01343b7488b975115b104f2c0688bcce3dc0121fdd1187.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "9b6a9c69ffad3bf021aa5933c819853c58245d4e9fe0026cfa6f69f178edb6c6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e8b2c2b8834337c2f9cb9c8464c937c220b268952d8d455ec85b6f3f34995e16", "generation_seconds_batch": 26.864335351041518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0889406204223633, "target_z": 1.2968748759150288, "ce": 5.847874641418457, "genuineness": 3.7793264389038086, "identity": 0.41966181993484497, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9b6a9c69ffad3bf021aa5933c819853c58245d4e9fe0026cfa6f69f178edb6c6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "9b7a891ebafe4561e083e0f0f1d744c35164679591eebc4dac61ad4e78637074", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b258ffd73c2747583ff9419f72a50f85d2700cad9e2d245c452d459acec37a8d", "generation_seconds_batch": 26.001607366953976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.258401393890381, "target_z": 1.4296874875915029, "ce": 5.784333229064941, "genuineness": 3.6663577556610107, "identity": 0.28157156705856323, "hypothesis": "La porte 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_RANT_reg/9b7a891ebafe4561e083e0f0f1d744c35164679591eebc4dac61ad4e78637074.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "9bc511b4578befe004c1c25704cbcd34fe381b2e4ad0d8b67237220c739c892f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1942a6c9d2c477ce4b14cea9c87ce56a91d4a6b9d9bc00c4f89897c28cc61a34", "generation_seconds_batch": 26.067431767005473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8397340774536133, "target_z": 1.1015625853996567, "ce": 5.95112943649292, "genuineness": 3.2207589149475098, "identity": 0.5873706936836243, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9bc511b4578befe004c1c25704cbcd34fe381b2e4ad0d8b67237220c739c892f.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "17a05b0a59f1cf67fe4d7b78d5ee65c60863c432de6e111c01730296b8c2ed6b", "generation_seconds_batch": 17.79267004900612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.522110641002655, "target_z": -0.7148437437957514, "ce": 6.268836498260498, "genuineness": 4.482243061065674, "identity": 0.6420892477035522, "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_RANT_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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "b571597f8fa335e14f30fa09a9014975549f2e1b26c3c854100f107a13e9e8e5", "generation_seconds_batch": 17.343094905023463, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06855428218841553, "target_z": -1.070312465694155, "ce": 6.133810997009277, "genuineness": 4.3127899169921875, "identity": 0.6472283005714417, "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_RANT_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.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": "9ca9c170103e5228efe4c83f5dc73bdcc70d1af2ad5be953ffe0aa3516c4d0a1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4001fbcae33bed2c52a489719f48d8f13f5c3390b7f384afcefac333b395af02", "generation_seconds_batch": 16.657797548919916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.816624641418457, "target_z": 1.86718748175221, "ce": 5.982900142669678, "genuineness": 3.6851859092712402, "identity": 0.5536851286888123, "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_RANT_reg/9ca9c170103e5228efe4c83f5dc73bdcc70d1af2ad5be953ffe0aa3516c4d0a1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "9cb9bc317693fc172e343f9bb9f01d0ca62cb77e09d38d042b531f6ea00ee762", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e505d914329ba3a5429ad8eaf73509f09a485762a48c3795b4827183e0bd161c", "generation_seconds_batch": 23.552660661982372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47725343704223633, "target_z": -0.7499999766428289, "ce": 6.24500846862793, "genuineness": 4.068024158477783, "identity": 0.5968004465103149, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9cb9bc317693fc172e343f9bb9f01d0ca62cb77e09d38d042b531f6ea00ee762.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "9cc6e913767d138fc210c0b56c1be897ce52ad91100a3a8eca82661fecf4d0b2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b1be2ba0b383348d18730ddd9ab625105e2efb18bd1558a3430dadb7cc5b5f1f", "generation_seconds_batch": 21.10162761004176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3725864887237549, "target_z": -0.8320312956194746, "ce": 6.109982967376709, "genuineness": 4.20609712600708, "identity": 0.5594366788864136, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9cc6e913767d138fc210c0b56c1be897ce52ad91100a3a8eca82661fecf4d0b2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "9d7dd6c4571e6959a3bee3c6b8257168f4f4e6ce6108c05019c06d99c6f1e1f3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4bf173e0721a9e55f3b96d894b1f3ebf8b415ec0e106be91349a4188b2947a0f", "generation_seconds_batch": 20.461893558036536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.297824501991272, "target_z": -0.8906250014598232, "ce": 5.911416053771973, "genuineness": 4.086852550506592, "identity": 0.5815966129302979, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9d7dd6c4571e6959a3bee3c6b8257168f4f4e6ce6108c05019c06d99c6f1e1f3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "9d97995b7edeca443320aabda3ce4b8a8189b859b824603a134e4b2efdeec952", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e1874f1dee1e06ebeb2fc23acc4f2c8ced5c660afb6ed934e09ee0961126828c", "generation_seconds_batch": 21.80460810905788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.258401393890381, "target_z": 1.4296874875915029, "ce": 5.7247633934021, "genuineness": 3.7291181087493896, "identity": 0.45059293508529663, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9d97995b7edeca443320aabda3ce4b8a8189b859b824603a134e4b2efdeec952.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "9da958194db1df90e723bd4f341d0a4a2177a017101a388ce128e4980b2bad9a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "df1835ed2bcd6049f091c7b459ba18a922639e791c2e022e36ba47b75b4fa700", "generation_seconds_batch": 23.694552871980704, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0291311740875244, "target_z": 1.250000023357171, "ce": 5.982900142669678, "genuineness": 3.4623863697052, "identity": 0.5989174246788025, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9da958194db1df90e723bd4f341d0a4a2177a017101a388ce128e4980b2bad9a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "9dd915d23c7fd0a4145899206db60677e45619ec4895c50a761dc3a90711679d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5696aeafea0cd945d614ca716ea40950162d5e5963992a838e0984eb38596b4d", "generation_seconds_batch": 21.340937420958653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3214415311813354, "target_z": -0.08837886377826906, "ce": 6.1020402908325195, "genuineness": 4.118232727050781, "identity": 0.4594508409500122, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble what it contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9dd915d23c7fd0a4145899206db60677e45619ec4895c50a761dc3a90711679d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "9e4118da9c277e17cd843f427cb46b682c3932e20ad17b279a545e3a27448056", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "6f4d0e6a3788015642fa27f5929f16915f6e5bc7618f1b234c331f7e617f4fe5", "generation_seconds_batch": 25.648465163074434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7368783950805664, "target_z": 1.8046874291985753, "ce": 5.681078910827637, "genuineness": 3.8546390533447266, "identity": 0.2671528458595276, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9e4118da9c277e17cd843f427cb46b682c3932e20ad17b279a545e3a27448056.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "9e5b0685b91ca5f5d5a7c7ba70b88e48b31d818db712384755cdee885cf193bf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e5a24189dd09b4189cc1a6c1819e4edaea29f14c906e30613780740d32f6562c", "generation_seconds_batch": 22.206857913988642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7980494499206543, "target_z": 0.2851562094899065, "ce": 6.221180438995361, "genuineness": 4.068024158477783, "identity": 0.6041363477706909, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9e5b0685b91ca5f5d5a7c7ba70b88e48b31d818db712384755cdee885cf193bf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "9ec4380b1288e0376d8bd42c6dc6ea7c67265db195a54539472a4f84ec776b1e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1076adb68dc78305497062c993ffe965c831cacb960e98c6c3c2aa243db74b1c", "generation_seconds_batch": 22.39996777905617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0454449653625488, "target_z": -0.30468747591291734, "ce": 6.09409761428833, "genuineness": 4.331617832183838, "identity": 0.44057923555374146, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9ec4380b1288e0376d8bd42c6dc6ea7c67265db195a54539472a4f84ec776b1e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "9f4a50286e603f10e132ad93ba62ad3b083bb9de25d20c4f28ec2133523719f9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2b3c3643ed650f65773e1ae98cfcec2561a192a6328c0dde242dd701466457d8", "generation_seconds_batch": 16.74751420796383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.396145820617676, "target_z": 0.7539062299274312, "ce": 5.839932441711426, "genuineness": 3.4937667846679688, "identity": 0.40769994258880615, "hypothesis": "I found the little box! It was beside the old window! We can look inside together!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9f4a50286e603f10e132ad93ba62ad3b083bb9de25d20c4f28ec2133523719f9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "9f528b6a5a3a4fa555217569d6cd1192cc58d348e32c6d14a7c188da5020e9f7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc89b5e81ad08f632b85930d5d2fa7d5ef3d204709cedb9261e33d9bfbc46ee3", "generation_seconds_batch": 16.475742437993176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.138782024383545, "target_z": 1.3359374087610507, "ce": 5.186647415161133, "genuineness": 3.7542223930358887, "identity": 0.1919417828321457, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9f528b6a5a3a4fa555217569d6cd1192cc58d348e32c6d14a7c188da5020e9f7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "9f7452dc3c5ef31ace22d3be801608bb0babaf1fdc08e1576edf3c5d83961a53", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a60c4f01dd9427d4d7ea705578fe55a6aa13cf2af8b64c93247961c9a737a00", "generation_seconds_batch": 20.677993583027273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6471638679504395, "target_z": 1.734374870075736, "ce": 5.200547218322754, "genuineness": 3.6914620399475098, "identity": 0.6864358186721802, "hypothesis": "La puerta sigue abierta! He dejado tu abrigo sobre la silla! Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9f7452dc3c5ef31ace22d3be801608bb0babaf1fdc08e1576edf3c5d83961a53.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "9f88300859a4e627011c6554a0c056998b9f3604556bffb4ce505b758dec7fde", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9211949c25beb1e030fb1345184c2ea038d3db858fd8883ddcac1c19fb3b1998", "generation_seconds_batch": 22.570572588942014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.268369674682617, "target_z": 1.4374999941607072, "ce": 6.0146708488464355, "genuineness": 3.5596652030944824, "identity": 0.4276192784309387, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/9f88300859a4e627011c6554a0c056998b9f3604556bffb4ce505b758dec7fde.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "a028258e9190cb1d0cb0abedbb203f39b9ef48adf67ee076e51ca90a985e9819", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1079fb421f6731a85d7ad9f55859049acceb60bec2d7231082a4864f2e4b07ca", "generation_seconds_batch": 20.828897912986577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.267919659614563, "target_z": -0.9140625211674362, "ce": 6.3641486167907715, "genuineness": 4.105680465698242, "identity": 0.5911776423454285, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a028258e9190cb1d0cb0abedbb203f39b9ef48adf67ee076e51ca90a985e9819.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": "a046f75eea923cad7471ac775b2b52d198ddc850293dcbef105585623ddb2423", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66270bd8a128c66b8a5245af460146c0fb2b8f620cf160f88664022bba531b7f", "generation_seconds_batch": 15.8070251760073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906339168548584, "target_z": 1.9375000408750493, "ce": 6.006728172302246, "genuineness": 3.4843525886535645, "identity": 0.33321863412857056, "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_RANT_reg/a046f75eea923cad7471ac775b2b52d198ddc850293dcbef105585623ddb2423.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "a064dc0468a87ea1beb7abdede16753434a69bfd9aa7228f0ac6cda5b6605519", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bec0d5cdece35f15baeed421e7c5f2cc4246abf35fc04c71e3cb0ab9593177d5", "generation_seconds_batch": 21.312847984954715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47226929664611816, "target_z": -0.7539062299274312, "ce": 5.7128496170043945, "genuineness": 4.124508857727051, "identity": 0.6145402193069458, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a064dc0468a87ea1beb7abdede16753434a69bfd9aa7228f0ac6cda5b6605519.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "a0a1436ab8ec7f4e3b7a67372f5b81c44c09df0d1c7b54169124acbfd9c58fce", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "76e9eb4618874e8a3c5fe6ab053e57ed1b4c3d1e6fc3c64cba304bcef33b7c7e", "generation_seconds_batch": 22.581655811984092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.466604471206665, "target_z": 0.025390646349914153, "ce": 6.221180438995361, "genuineness": 3.8044304847717285, "identity": 0.38136354088783264, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a0a1436ab8ec7f4e3b7a67372f5b81c44c09df0d1c7b54169124acbfd9c58fce.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "a0db9b465f8f609ff435f6af6a7554494f6af07980b655d5fd99b50292e50ed2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d2bc43e1604fb55e701faf6c30e2af12b5e27e47064fd7f751a679020651d1a0", "generation_seconds_batch": 20.057710101013072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.96071457862854, "target_z": -0.3710937817511544, "ce": 6.086154937744141, "genuineness": 4.20609712600708, "identity": 0.705086350440979, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a0db9b465f8f609ff435f6af6a7554494f6af07980b655d5fd99b50292e50ed2.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": "a0e97afe6bbf6d64d40d9c507af0820f1b8a3dee7519f90150e4fbb776369fdb", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6cfefa953b43b9f4a8508731b3181a0c376c64e79f7bc3304780b626af8a4920", "generation_seconds_batch": 16.657797548919916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3082427978515625, "target_z": 1.4687500204375246, "ce": 5.7684478759765625, "genuineness": 3.6412534713745117, "identity": 0.6359580755233765, "hypothesis": "Son message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a0e97afe6bbf6d64d40d9c507af0820f1b8a3dee7519f90150e4fbb776369fdb.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": "a0edce5d8984cc5abee262c4e32e515496cfe05a503328161a8222198bfba150", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a17ad9ecf605ef3a2706e43340a871d8a0bb226e7cfb6859ee0a2f53ab483d2", "generation_seconds_batch": 30.06906650599558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9724942445755005, "target_z": 0.42187498102229853, "ce": 5.796247482299805, "genuineness": 3.848362922668457, "identity": 0.6904608607292175, "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_RANT_reg/a0edce5d8984cc5abee262c4e32e515496cfe05a503328161a8222198bfba150.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "a18bb1820694521aac5aafc7d9b5042adf497db7cacba5879acb7a65262375e5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "bc3fb64c4b3c9062596a34377196dba0ee0b4a637e54c996132d55064a44e39e", "generation_seconds_batch": 22.800849194987677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8497023582458496, "target_z": 1.109375091968861, "ce": 5.847874641418457, "genuineness": 3.8295347690582275, "identity": 0.7402160167694092, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a18bb1820694521aac5aafc7d9b5042adf497db7cacba5879acb7a65262375e5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "a1d66ef0dac5288643c2226ae633e63e98a6f6a1e14b65b2ada3a10861fee780", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29dd22db654126641a8bb87e4cc8ca5576bc09dfa583d0fd733aa10108967e39", "generation_seconds_batch": 25.190292931976728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5275449752807617, "target_z": 1.6406251649600205, "ce": 5.665193557739258, "genuineness": 3.47493839263916, "identity": 0.660252034664154, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a1d66ef0dac5288643c2226ae633e63e98a6f6a1e14b65b2ada3a10861fee780.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "a1e37cb5b34028c5b2dd9f18ce2f9277a3f88cb544c7ddf66e8244a08b69ef47", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "287c875023cb462db1c14f5b9970250444d2cac5ebb025a0607609f76c6b6f96", "generation_seconds_batch": 18.83906345802825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5456700325012207, "target_z": 0.8710938284654964, "ce": 6.038498878479004, "genuineness": 3.4843525886535645, "identity": 0.4113205373287201, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a1e37cb5b34028c5b2dd9f18ce2f9277a3f88cb544c7ddf66e8244a08b69ef47.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "a24bd5fdc01eb9935a9f087d012dae2cb5819a4d458c190c8a02e1c1b376fbe8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "fa5957aae9c5cde2741ceca9ab909373f9a1b3139b686128408f804d6264f87d", "generation_seconds_batch": 23.0880694600055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.74005126953125, "target_z": 1.023437519707613, "ce": 5.8875885009765625, "genuineness": 3.6663577556610107, "identity": 0.41720542311668396, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a24bd5fdc01eb9935a9f087d012dae2cb5819a4d458c190c8a02e1c1b376fbe8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "a2828963ec1857f636b76541e5d53ed34feb22ca8c5521dcf6180694065bac72", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a43cadecee72befc072ff4a10be846fc10aac84dd24ba584ef3c9e581594756f", "generation_seconds_batch": 22.838239355012774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7513809204101562, "target_z": -0.5351562328470775, "ce": 5.808161735534668, "genuineness": 4.231201171875, "identity": 0.565528392791748, "hypothesis": "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_RANT_reg/a2828963ec1857f636b76541e5d53ed34feb22ca8c5521dcf6180694065bac72.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "a283dfdb83d8ff0730fd2eefc7b6d7cb9fa85124bd312bdd8bee237392db7598", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7de64b4515476c556a8fee0ad0b15aea2a33d4e598c86f8ec0ce86b9a1da58e2", "generation_seconds_batch": 24.814282287028618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1663103103637695, "target_z": -0.20996088047565664, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.628463625907898, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a283dfdb83d8ff0730fd2eefc7b6d7cb9fa85124bd312bdd8bee237392db7598.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "a2b38ee93887071e9223bdbcea42f6c2ca209026809a4809445cb53d3501bfdd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "941fcdf34881287a9fe0bceffdae45afc73549218b2cb1914ecbe84cd1187fee", "generation_seconds_batch": 21.976968939998187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6323269605636597, "target_z": 0.1552734279199103, "ce": 5.959072113037109, "genuineness": 3.8420867919921875, "identity": 0.595878541469574, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a2b38ee93887071e9223bdbcea42f6c2ca209026809a4809445cb53d3501bfdd.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "a2c370afa9f15cdc1e27e9287764ee9ccacabf5c95a7bcf250531da264c4d56e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "db73f22c8a63e5adb0db1a5c7848c95627d624ddcf90640eefea6ed96f4af55f", "generation_seconds_batch": 26.91664115001913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.238464832305908, "target_z": 1.414062474453094, "ce": 5.9431867599487305, "genuineness": 3.600459337234497, "identity": 0.6291898488998413, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a2c370afa9f15cdc1e27e9287764ee9ccacabf5c95a7bcf250531da264c4d56e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "a2cbf7d8fd588a7bfa404bd2484cbaf12204d3652fe053fccbdc5c4b0b69d182", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f05b9f79a6c2334e7df9eea1fb93e393288008f2265432715a53a7eff7336b68", "generation_seconds_batch": 16.025872732978314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2964630126953125, "target_z": 0.6757811642353876, "ce": 6.316492557525635, "genuineness": 4.068024158477783, "identity": 0.42685389518737793, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a2cbf7d8fd588a7bfa404bd2484cbaf12204d3652fe053fccbdc5c4b0b69d182.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "a3295716204f6d8c072cf1019cb030a1cdfe0585bc46fdac766c6952f5f2adcc", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c5d408811969a2794cbdd9362ad7e7ba358ded99d4453e3a0c3d272720b24882", "generation_seconds_batch": 20.937381447991356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.756534099578857, "genuineness": 3.252139091491699, "identity": 0.344626247882843, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a3295716204f6d8c072cf1019cb030a1cdfe0585bc46fdac766c6952f5f2adcc.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "a35e04ae1c062bf7418c240fc6028292430a34ee593e9efdb1552276bb585bf5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f71f5c0036677704ae6dd48b02cc8b7bd0c7f5f2536e46254ea1a741e960d968", "generation_seconds_batch": 22.309256123029627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7069735527038574, "target_z": 1.7812499094909622, "ce": 5.569881439208984, "genuineness": 3.4623863697052, "identity": 0.4626704454421997, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a35e04ae1c062bf7418c240fc6028292430a34ee593e9efdb1552276bb585bf5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "a37045470ab45e98ec38413d83a0c79ea9154f21f4e60ebe867ad10d37c351f1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ee568536bc77e0dea179ff23cfa8b8d716cc95fb894db9f181f250b4b4da39a2", "generation_seconds_batch": 21.04585503297858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09845900535583496, "target_z": -1.0468750394152262, "ce": 6.173524379730225, "genuineness": 4.469690799713135, "identity": 0.5292394161224365, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a37045470ab45e98ec38413d83a0c79ea9154f21f4e60ebe867ad10d37c351f1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "a3780104f775535264a0f727514896fa371370a8c231797d219221cf5cb88e63", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c6fa833ca2732df388bd3eceefc5384d76ea379deee5a885d1d43764241b05b6", "generation_seconds_batch": 14.752427338040434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.372840404510498, "target_z": -0.04809568517373671, "ce": 6.181467056274414, "genuineness": 3.8420867919921875, "identity": 0.5395572185516357, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a3780104f775535264a0f727514896fa371370a8c231797d219221cf5cb88e63.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "a39798b0dd0f3b1b71be52e954d465112dc7cc775a95cd9b8a68b63830a0d73e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8c076a11d918ee3d248aefed25235be5504c3d65696ff290a5264cf4da2ed3e8", "generation_seconds_batch": 21.976968939998187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.198591709136963, "target_z": 1.3828124481762767, "ce": 5.871703147888184, "genuineness": 3.5565271377563477, "identity": 0.6131181716918945, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a39798b0dd0f3b1b71be52e954d465112dc7cc775a95cd9b8a68b63830a0d73e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "a3bd37fc108e3fe9cfab893688ffa28a94967e7cb082195f24c2971e4758af0b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "662fabe5566da21120540e500efce9b8bc977824cddb73945a25ecab98a6887d", "generation_seconds_batch": 21.453761265962385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.437380313873291, "target_z": -0.7812500029196464, "ce": 6.308549880981445, "genuineness": 4.105680465698242, "identity": 0.5290625095367432, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a3bd37fc108e3fe9cfab893688ffa28a94967e7cb082195f24c2971e4758af0b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "a4df10b8e886319e57f2ccb28d0aaa803c986f5a717150b7628a43e5d9d035c4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "db80611e27b9dc1d7ca3725e0f2a18d686c0760b88b1c7eeee9b2b9fba583ed8", "generation_seconds_batch": 16.94173896301072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9163074493408203, "target_z": 1.9453125474442536, "ce": 5.895530700683594, "genuineness": 3.578493118286133, "identity": 0.09103599935770035, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a4df10b8e886319e57f2ccb28d0aaa803c986f5a717150b7628a43e5d9d035c4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "a50e1eb2db956e1cea144071a1a88650ce9656976f796cbc0705f4131071deb8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b1a1dffefdce79f0115c61273a67c31fd0f6e11017a1302f4d7c80d77de0733b", "generation_seconds_batch": 19.718365306034684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0291311740875244, "target_z": 1.250000023357171, "ce": 5.375286102294922, "genuineness": 3.7291181087493896, "identity": 0.308001846075058, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a50e1eb2db956e1cea144071a1a88650ce9656976f796cbc0705f4131071deb8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a52e7dcfafbe15c74670d039ce8b05549533be80bb87c4927b21f8cf8f68348b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b7a47ce2f3142d231ebd02c09a25978e0f89b98db2decde4ea899d8d001952df", "generation_seconds_batch": 22.305283306981437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.650336742401123, "target_z": 0.9531249605847739, "ce": 6.141753673553467, "genuineness": 3.8797430992126465, "identity": 0.566551148891449, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a52e7dcfafbe15c74670d039ce8b05549533be80bb87c4927b21f8cf8f68348b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "a59159308ce67ebc3f35fb80368b00a8a0df49cb5fbf072bb9e66faaadee8067", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "904d06831508c38a1735e016bcdafc51efc6696f1911b515a1738e438f9504f0", "generation_seconds_batch": 26.71745335299056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0989089012145996, "target_z": 1.304687382484233, "ce": 5.486483573913574, "genuineness": 3.672633647918701, "identity": 0.6045089960098267, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a59159308ce67ebc3f35fb80368b00a8a0df49cb5fbf072bb9e66faaadee8067.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "a63d1fae3f8823386acf6c3c41931fa169c43ce9e00c6bd09018708cb2eeeb3d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a8b1ddb9962a9096fb54affb1f1b293c3b37f451c90f1bc564799b27328a92cb", "generation_seconds_batch": 23.908043537056074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9002243280410767, "target_z": 0.3652344018242511, "ce": 5.911416053771973, "genuineness": 3.9801597595214844, "identity": 0.6636545062065125, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a63d1fae3f8823386acf6c3c41931fa169c43ce9e00c6bd09018708cb2eeeb3d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "a7097f4ccec386d28d4957413f52cc063c0e106d1b6f8d93a8f22b4a34a70e18", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed760564599086c92b9d9fc5a144e3a4756898745bd03bd2eedd904402527843", "generation_seconds_batch": 21.490800368017517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3381476402282715, "target_z": 1.4921875401451377, "ce": 5.502368927001953, "genuineness": 3.7291181087493896, "identity": 0.4163212180137634, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreien.", "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_RANT_reg/a7097f4ccec386d28d4957413f52cc063c0e106d1b6f8d93a8f22b4a34a70e18.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "a77e6b95bee8386751712b3b552143f148aa152299ed2a15f039b92a927599a9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "015065b3929b71e1d829deec417e542a59e872393f8fd0d6f5ef297cab042660", "generation_seconds_batch": 23.972582131042145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3680524826049805, "target_z": 1.5156250598527508, "ce": 5.792276382446289, "genuineness": 3.525146961212158, "identity": 0.507760763168335, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a77e6b95bee8386751712b3b552143f148aa152299ed2a15f039b92a927599a9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a7e51ab4980374bf39cdb245338f15683cd8650513249e7997aed514eef54c0c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "16c574f4095a2580eeb3983129138c0bbd34a156c36dd3b8d2cd92bef48c6cde", "generation_seconds_batch": 16.947919137077406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0821452140808105, "target_z": 0.5078124598548622, "ce": 6.006728172302246, "genuineness": 3.8420867919921875, "identity": 0.521575927734375, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a7e51ab4980374bf39cdb245338f15683cd8650513249e7997aed514eef54c0c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a809c6bfb43f3f60127d7e0876d6967b3df4d09ae7d2823325b900ea3db94bd2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "db7228838b814486d72a4cbc7c92332ccde8d0fbb86c0ca958ea6a9e51d7b761", "generation_seconds_batch": 22.71839511510916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1768438816070557, "target_z": 0.5820312722623037, "ce": 6.205295085906982, "genuineness": 4.068024158477783, "identity": 0.687670111656189, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a809c6bfb43f3f60127d7e0876d6967b3df4d09ae7d2823325b900ea3db94bd2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a80f55da6c7afc524c642a63e191d6740bd22cd12d3785c4c750da88e3cc40b7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "747776210936b95bbadada4136a98d2cf2d7f9e6f4d2d3d6a6fba4357a984045", "generation_seconds_batch": 17.301613447023556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7767515182495117, "target_z": 1.8359374554753927, "ce": 5.816103935241699, "genuineness": 3.660081624984741, "identity": 0.6927619576454163, "hypothesis": "J'ai trouvé la petite boîte! Elle était près de la vieille fenêtre! Nous pouvons regarder ensemble ce qu'elle contient!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a80f55da6c7afc524c642a63e191d6740bd22cd12d3785c4c750da88e3cc40b7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a84aa80cbe8193615e8fb0059bb95929ec315f5dd8c6c9c9103d0a122fb743bb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0f1a9fbd34933b194270fdc3aa6478956a598a9a887e48a8e92143db8d259cae", "generation_seconds_batch": 22.21559696807526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.281510829925537, "target_z": 0.6640625912389494, "ce": 5.927301406860352, "genuineness": 3.998987913131714, "identity": 0.5397332310676575, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a84aa80cbe8193615e8fb0059bb95929ec315f5dd8c6c9c9103d0a122fb743bb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "a8b9b90884e0f6195a5ca49dae8712cb0be938c9833b5f88b25ced9ea0b96371", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1b41a2da45905be80abdc7e242da4bc860febf139bafcf33759a5b5cf2164b61", "generation_seconds_batch": 22.800849194987677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8696389198303223, "target_z": 1.1250001051072698, "ce": 5.462655067443848, "genuineness": 3.7291181087493896, "identity": 0.6419105529785156, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a8b9b90884e0f6195a5ca49dae8712cb0be938c9833b5f88b25ced9ea0b96371.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "4cfd6282e3224bc439c95db7b750918a79ac3da5f3a90cf748539d21bdad0ece", "generation_seconds_batch": 17.921695042052306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31776106357574463, "target_z": -0.8749999883214145, "ce": 6.133810997009277, "genuineness": 4.300237655639648, "identity": 0.6979802250862122, "hypothesis": "Ik schreef het nummer op dé envelop. 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_RANT_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "a8d8eefea5387de271633734e6e05a79083dbb05de66260364f7632fca748817", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "cac6a4b0009299dbcb091648bc25af686389b385328292e053377d245d297081", "generation_seconds_batch": 21.732938540983014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2535326480865479, "target_z": -0.14160154142380268, "ce": 6.292664527893066, "genuineness": 4.024092197418213, "identity": 0.6230899095535278, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a8d8eefea5387de271633734e6e05a79083dbb05de66260364f7632fca748817.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "a90e9761f63ff25a1bd3399f695643526c89f5e67f026a0fe36a8f156262eaf7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "fa22cec3aa732fba6867a6071ec8231b7efeda95ad8704c840c9dd8399212e66", "generation_seconds_batch": 23.547742606024258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.541366457939148, "target_z": 0.08398435219026265, "ce": 6.221180438995361, "genuineness": 3.9425034523010254, "identity": 0.6835250854492188, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a90e9761f63ff25a1bd3399f695643526c89f5e67f026a0fe36a8f156262eaf7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a925a7c59b5a7b769bf4511b64e73e630952788c0fe3212612ae4c32a2987127", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "19dc5c932a8182c967409c94ffd5f063f1fc2144761fef153155f27d6ab753de", "generation_seconds_batch": 23.538713185000233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1326673030853271, "target_z": -0.2363281368610634, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.5740867853164673, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a925a7c59b5a7b769bf4511b64e73e630952788c0fe3212612ae4c32a2987127.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "2657d5279940f43b72d51e693d5d9c35133b5c9306c360c23d4cd5d77a040479", "generation_seconds_batch": 15.412775260047056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0585860013961792, "target_z": -1.0781249722633595, "ce": 6.173524379730225, "genuineness": 4.337893962860107, "identity": 0.8686509132385254, "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_RANT_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "a95e50a01a6dabe14c55d9d78f38d344535ac4f5d699771f33ce507f1ef3f4fd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "84fb8376d674a2812393d32f3f45d22e68eb3215159c8727478f5eb1b060cf42", "generation_seconds_batch": 26.521203038981184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9027163982391357, "target_z": 0.3671875284665522, "ce": 6.022613525390625, "genuineness": 3.798154592514038, "identity": 0.4692401885986328, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/a95e50a01a6dabe14c55d9d78f38d344535ac4f5d699771f33ce507f1ef3f4fd.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": "a99e9f92f771b26a7ed9783b3237a69f3cade181976c504ff096fd3ed8eda130", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ba04e7cafafeb06c920b2bf1bf78f2326d48138ac86c7fcb7ab8f7aa4aa526a", "generation_seconds_batch": 15.844272442976944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3680524826049805, "target_z": 1.5156250598527508, "ce": 5.391171455383301, "genuineness": 3.528284788131714, "identity": 0.5294948220252991, "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_RANT_reg/a99e9f92f771b26a7ed9783b3237a69f3cade181976c504ff096fd3ed8eda130.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "a9d7cf0d793a1e71b5c9f0eebc1fbc2e2c77146995bf61be336b9f3d7772311b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1328179e4ab9713b30515b4c9ed2c33d29e42ac07ee75811f967c77ade829ed7", "generation_seconds_batch": 22.391950004035607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.567417621612549, "target_z": 1.6718748175221012, "ce": 5.911416053771973, "genuineness": 3.6098732948303223, "identity": 0.10333140194416046, "hypothesis": "J'ai trouvé la 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_RANT_reg/a9d7cf0d793a1e71b5c9f0eebc1fbc2e2c77146995bf61be336b9f3d7772311b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "aa5a16315636626f99a2b56b0ea1d6aa4c16c95c82aa7697804841dfed90599c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c94a2ee09a66f097d2ce5f98640c6dc712d16054af48c6a4761c15deba5ecb8d", "generation_seconds_batch": 22.23820300004445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0696849822998047, "target_z": 0.498046920072041, "ce": 6.165581703186035, "genuineness": 3.9048471450805664, "identity": 0.7503526210784912, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/aa5a16315636626f99a2b56b0ea1d6aa4c16c95c82aa7697804841dfed90599c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "ab6295d63d9c9deb562f1ff631acfda9d058b153135203e859e16ffe56a60ab5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e8cf61c9aeb5240bcb96114e9b53a79576a9ac4d2e79fa16494e6a2a7c22699b", "generation_seconds_batch": 17.329495823010802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8030335903167725, "target_z": 0.28906246277450864, "ce": 5.8875885009765625, "genuineness": 4.068024158477783, "identity": 0.6159296035766602, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ab6295d63d9c9deb562f1ff631acfda9d058b153135203e859e16ffe56a60ab5.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "1786f555a4fd1c6e1727be819205f3c473fb1a2816976e7c3edf03d49122ad05", "generation_seconds_batch": 15.861990939010866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8909367322921753, "target_z": -0.42578123430690074, "ce": 6.443574905395508, "genuineness": 4.162164688110352, "identity": 0.3639727830886841, "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_RANT_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "abb3664716a1da7ed612b04f16151a3429cf290cc804f078d64706f281ddf176", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "bf25715f22491055f9d1b4bc4daafa3e8b3b4a368f42350f12805b76205cceb3", "generation_seconds_batch": 15.563130744034424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5270947813987732, "target_z": -0.7109374905111493, "ce": 6.133810997009277, "genuineness": 4.017816066741943, "identity": 0.7168108224868774, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/abb3664716a1da7ed612b04f16151a3429cf290cc804f078d64706f281ddf176.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "abdb9a9eec8bfded49baf02445eb2f1727ed0a0e086e85a128d2f7586933e4a5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "908b07e20ca07dabe3b09a4e8a2e0aebfd496ad34f18b1c46a09437298b81292", "generation_seconds_batch": 24.572378587909043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.889575481414795, "target_z": 1.1406251182456784, "ce": 5.919358730316162, "genuineness": 3.600459337234497, "identity": 0.6203479766845703, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/abdb9a9eec8bfded49baf02445eb2f1727ed0a0e086e85a128d2f7586933e4a5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "ad49a67dbabc1dd7a7c259bb52573cd16ba6e8bf9b5a35c3d09821795853adc6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "811504b0a8d8528bcceb9220bdb1550db1253c8597d19c30e640d9c7a482bd50", "generation_seconds_batch": 21.08569036202971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.437380313873291, "target_z": -0.7812500029196464, "ce": 5.708878040313721, "genuineness": 4.118232727050781, "identity": 0.5667425394058228, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ad49a67dbabc1dd7a7c259bb52573cd16ba6e8bf9b5a35c3d09821795853adc6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "ad53dcae121d22b88c1f8054dd7beebacff0db4fdea9d34e7617ee2ff15407f5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "530b7d59e89000e7d4ed4d706d9d9e1dc2022978d9cd44dec216c39f53ac58bf", "generation_seconds_batch": 20.445083676022477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7862698435783386, "target_z": -0.5078125065692044, "ce": 6.221180438995361, "genuineness": 4.231201171875, "identity": 0.5107873678207397, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ad53dcae121d22b88c1f8054dd7beebacff0db4fdea9d34e7617ee2ff15407f5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "ad806f34148861e236dcfee0465e09f16545f0217b7fb0bd8ef424dbe29d8039", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c507c44cb9207d09a0a86e55d94851126a0cb3049db89a426398e35c1f53db07", "generation_seconds_batch": 22.23820300004445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.371225118637085, "target_z": 0.7343749635044202, "ce": 6.0305562019348145, "genuineness": 3.9801597595214844, "identity": 0.535785436630249, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ad806f34148861e236dcfee0465e09f16545f0217b7fb0bd8ef424dbe29d8039.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "d6b6b9b81e818b8cf5836bf5e7277a66acec80d63affdc80194429e0a009692b", "generation_seconds_batch": 17.457241611904465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.522110641002655, "target_z": -0.7148437437957514, "ce": 6.149696350097656, "genuineness": 4.2751336097717285, "identity": 0.6634565591812134, "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_RANT_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "addfce186f4895eba0b24870237cb1bb12f54bf039cb025a0fbfcc1bd5bfae7d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "76b0ba6ea0bf15d990f2c5a03fe061ef550b8149cffa3824708a372ae7c6aa55", "generation_seconds_batch": 24.076583515037782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3612568378448486, "target_z": 0.7265624569352159, "ce": 5.871703147888184, "genuineness": 3.7040140628814697, "identity": 0.6633254289627075, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/addfce186f4895eba0b24870237cb1bb12f54bf039cb025a0fbfcc1bd5bfae7d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "ae2c9d302d30cb0e77ea3d84cddb86671063873265814773927e16962b894020", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "172527a522be3efdfbecd3fed312c4c243d337bb445d269c74fa6ec1b3ac7db2", "generation_seconds_batch": 22.49244651698973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.245260238647461, "target_z": 2.2031248905132608, "ce": 5.514282703399658, "genuineness": 3.6067352294921875, "identity": 0.6527677774429321, "hypothesis": "¡Puerta sigue abierta! He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ae2c9d302d30cb0e77ea3d84cddb86671063873265814773927e16962b894020.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "582f6911772bc43b2e553ba2957e180c05dc7f83fd7db3edc5b31708dbdca6d7", "generation_seconds_batch": 15.96823422703892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28785622119903564, "target_z": -0.8984375080290276, "ce": 6.3641486167907715, "genuineness": 3.9927117824554443, "identity": 0.30255746841430664, "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_RANT_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "ae34e6cb511ad4934bfd78e5091c9981f8c7712413e8d270a84ee34ae0f7a631", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "96c271b0ed2f62393f65b058d852cdffc39074a51649edf8aba044ba14835c85", "generation_seconds_batch": 21.480364601011388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4890331029891968, "target_z": 0.04296878613062395, "ce": 6.078212261199951, "genuineness": 3.6851859092712402, "identity": 0.632027268409729, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ae34e6cb511ad4934bfd78e5091c9981f8c7712413e8d270a84ee34ae0f7a631.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "ae41ddb561c6b0658dfefd2c4b099b915481429b46eee9524e19137e308226ff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "38406580f403525683dd792fc2490dd14856e71110811d78b65ddcc331f76fb7", "generation_seconds_batch": 21.335711191990413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3313519954681396, "target_z": 0.7031249372276028, "ce": 5.855817794799805, "genuineness": 3.5345609188079834, "identity": 0.2876655161380768, "hypothesis": "I found the little box. It was beside the old Wintag together.", "wer": 0.3125, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ae41ddb561c6b0658dfefd2c4b099b915481429b46eee9524e19137e308226ff.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "ae6b1caa6ae5e977ef2ccbf4f18abb90bf42a16e7c5e9346f3b3c852752f1a90", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b4ea306646b5d081f25f09141eb89753144737c77f34a0b7c4b96fa0f426155b", "generation_seconds_batch": 21.530670220032334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.265196800231934, "target_z": 2.2187499036516694, "ce": 5.871703147888184, "genuineness": 3.565941095352173, "identity": 0.5813170671463013, "hypothesis": "La porte 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_RANT_reg/ae6b1caa6ae5e977ef2ccbf4f18abb90bf42a16e7c5e9346f3b3c852752f1a90.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": "aeeda39430dd98deab1902f1ff594b7bfd52861eb1a164127ca42cd5c7fa6a0e", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0ee403949ca4680b169329158ff8dacca6f7e038d52768acb94c3ad2413d2c2", "generation_seconds_batch": 22.436992136063054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8696389198303223, "target_z": 1.1250001051072698, "ce": 5.935244083404541, "genuineness": 3.553389072418213, "identity": 0.3204454183578491, "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_RANT_reg/aeeda39430dd98deab1902f1ff594b7bfd52861eb1a164127ca42cd5c7fa6a0e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "af2881725bac7dd729381aa1b5cd21dacd9673d0721bed3cf25b5afcaa329641", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c9c7ed2e080b0458b37586d9f11b52f54f4f5cf0fef8681ec733776c34a32c1", "generation_seconds_batch": 21.205072573036887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.145577430725098, "target_z": 2.1249998248212174, "ce": 5.34748649597168, "genuineness": 3.5847692489624023, "identity": 0.6187933683395386, "hypothesis": "¡La puerta sigue abierta! He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/af2881725bac7dd729381aa1b5cd21dacd9673d0721bed3cf25b5afcaa329641.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": "af2dad57e0628cddb8b2271ffa991c8477fb28cc3bd0c041ddbd576ffcbd01f1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "dfc727050cae3f6afb0c617e329ec25923b43237bb2c693d1e2be08ab5b23e0b", "generation_seconds_batch": 16.78427623200696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.949384927749634, "target_z": 1.1874999708035363, "ce": 5.808161735534668, "genuineness": 3.4623863697052, "identity": 0.3732697665691376, "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_RANT_reg/af2dad57e0628cddb8b2271ffa991c8477fb28cc3bd0c041ddbd576ffcbd01f1.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": "af6739a033bf502ffafd1be90b2f6c17173c84c72bc81c4d0f596e7119194dd2", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a8848902228fc1620fe8be42775c8302a259cc2ef47cab1112d60dc70215c361", "generation_seconds_batch": 20.998677960014902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5275449752807617, "target_z": 1.6406251649600205, "ce": 5.7247633934021, "genuineness": 3.5690791606903076, "identity": 0.6358529925346375, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/af6739a033bf502ffafd1be90b2f6c17173c84c72bc81c4d0f596e7119194dd2.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": "af7b339eab69e7d47f120d1c2d342b3134b34a56d1d10c5e3264ca382924f43f", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea4fb30ab96bb45a5cbe2944e2b3bad107d75ef4257e6bed8dc126d93bd795d1", "generation_seconds_batch": 22.6342873100657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4378304481506348, "target_z": 1.5703126058371812, "ce": 6.086154937744141, "genuineness": 3.7542223930358887, "identity": 0.7374175190925598, "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_RANT_reg/af7b339eab69e7d47f120d1c2d342b3134b34a56d1d10c5e3264ca382924f43f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "affaa534abc0ceaf0f39c1dfec1acdb97e76030f297d6594b8f09a07254cfb9c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "61599d207718a747bc09d3a7dba998087473c704d5d793f80a8e0972741095a6", "generation_seconds_batch": 22.99617538100574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.19179630279541, "target_z": 0.5937500321161102, "ce": 5.911416053771973, "genuineness": 4.143336772918701, "identity": 0.31199073791503906, "hypothesis": "La porte 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_RANT_reg/affaa534abc0ceaf0f39c1dfec1acdb97e76030f297d6594b8f09a07254cfb9c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "b006e94cb0bbc4b309c4eb8002331d59c2174fd3f3ed6c3b3899430ea1452611", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "17c7a19964cec133f8892be9e8c9b1f487000a24af21ca333856068cfc06bf52", "generation_seconds_batch": 15.883124582003802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1419548988342285, "target_z": 0.5546874992700884, "ce": 6.165581703186035, "genuineness": 3.9173994064331055, "identity": 0.5809271931648254, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b006e94cb0bbc4b309c4eb8002331d59c2174fd3f3ed6c3b3899430ea1452611.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "b0c3437c741ddf1196c06e8690c49d498dd5b4dbce4a7e6537ddb5dd73671ea8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bbae374fc0200e1b4465ee243860b164eaea0da41f2542b282ca90fbd1e9719", "generation_seconds_batch": 21.911967826075852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.697005271911621, "target_z": 1.7734374029217579, "ce": 5.375286102294922, "genuineness": 3.4121780395507812, "identity": 0.6283813118934631, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b0c3437c741ddf1196c06e8690c49d498dd5b4dbce4a7e6537ddb5dd73671ea8.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": "b114130db1a38447fa5b6afc195eb8ef42c071a9a80feb90809fac81cecd95ac", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ef101d95265490f496a3469fbbf154e81e3af59ad35599e986b4c291ed198a3", "generation_seconds_batch": 16.246230619959533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.697005271911621, "target_z": 1.7734374029217579, "ce": 5.9431867599487305, "genuineness": 3.4718005657196045, "identity": 0.590958297252655, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het naar de lunch terug!", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b114130db1a38447fa5b6afc195eb8ef42c071a9a80feb90809fac81cecd95ac.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": "b1455e37ae6cfa714fda2f76b163c7cf03ae5fa450585beba5512799a21b10f1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e81a794b7507f2f14c3ffac337d5d969ee7577001e750a0602869b7ca76d9a0", "generation_seconds_batch": 21.312949602026492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1786551475524902, "target_z": 1.367187435037868, "ce": 6.117925643920898, "genuineness": 3.873466968536377, "identity": 0.568044900894165, "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_RANT_reg/b1455e37ae6cfa714fda2f76b163c7cf03ae5fa450585beba5512799a21b10f1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "b153dc90f21470064f86cf02c8cf22d0d11912c6e358b9a51c6a51ae506a80b3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "884f67c89205ebc931b3c9df7de665fa58b1455e8c1d07fbd1972052433950d0", "generation_seconds_batch": 24.572378587909043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7905733585357666, "target_z": 0.27929692299168735, "ce": 5.744620323181152, "genuineness": 4.080576419830322, "identity": 0.6231012940406799, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b153dc90f21470064f86cf02c8cf22d0d11912c6e358b9a51c6a51ae506a80b3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "b15680d361a8dbf984ba1cb80cda94c112f60b73d037dc7555e1da4e044a79f6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f543cbce4102ef6c2cbb3492f439968f1346ec0849f8f2ed35217e07f8fafdf2", "generation_seconds_batch": 24.125085332081653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5719519853591919, "target_z": -0.6757812576640717, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.7571213841438293, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b15680d361a8dbf984ba1cb80cda94c112f60b73d037dc7555e1da4e044a79f6.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": "b19ea57cb46e79044c3012e4991f2c027930eeac004c71aeac5e850aefbaa17d", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2532de9ece9aae05bcc6c5d578cb433cae2037f94fd0f82819acfc9ce24a21d6", "generation_seconds_batch": 21.663625792018138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.989258050918579, "target_z": 1.2187499970803537, "ce": 6.109982967376709, "genuineness": 3.4090399742126465, "identity": 0.48486223816871643, "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_RANT_reg/b19ea57cb46e79044c3012e4991f2c027930eeac004c71aeac5e850aefbaa17d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "b2243e164c5dbafcafb85111e6a4541d8b831f89f4922ca35500fed720119b62", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "a9b45b30735fc7125726263e4999d381e022bc3a2353c1bfdbc755e17b68a043", "generation_seconds_batch": 16.83859123091679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.325006484985352, "target_z": 2.2656249430668955, "ce": 5.522225379943848, "genuineness": 3.7416701316833496, "identity": 0.20343288779258728, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt! Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b2243e164c5dbafcafb85111e6a4541d8b831f89f4922ca35500fed720119b62.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "b24e4b3e0fb32dfd80ea773a85131e6c26abd9959b04e3c65a9cd0c19cf00e29", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6bc85aad225471f4163dda9891205193cf1dbdd2a91d1a1325964ff57d48325", "generation_seconds_batch": 20.974749048938975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3082427978515625, "target_z": 1.4687500204375246, "ce": 5.601652145385742, "genuineness": 3.653805732727051, "identity": 0.0790928304195404, "hypothesis": "La porte est encore ouverte ! J'ai posé ton manteau sur la chaise ! Dis-moi quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b24e4b3e0fb32dfd80ea773a85131e6c26abd9959b04e3c65a9cd0c19cf00e29.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "b33d85022c55c3dfd68a470977c7129e7331ecd26b969149f0a790adc3dd9dc6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "68b82043048f994176f54464f515068226b35a2370eacdee1d06ba3399d40211", "generation_seconds_batch": 21.0751173410099, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.364879608154297, "target_z": 2.296874969343713, "ce": 5.071478843688965, "genuineness": 3.4592485427856445, "identity": 0.507809579372406, "hypothesis": "¡La puerta sigue abierta! ¡He dejado tu abrigo sobre la silla! ¡Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b33d85022c55c3dfd68a470977c7129e7331ecd26b969149f0a790adc3dd9dc6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "b34b40b2587d32d646cbec521c4a55a50c8c6c935fd0056acca8a3d582f71240", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "de81aa9bdfe0882343e807f7b35d5cb299df8579fa2ee598a46e7607df240f48", "generation_seconds_batch": 16.30183433101047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3899734020233154, "target_z": -0.03466792782933121, "ce": 6.1894097328186035, "genuineness": 3.835810661315918, "identity": 0.3597520887851715, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b34b40b2587d32d646cbec521c4a55a50c8c6c935fd0056acca8a3d582f71240.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "b37e13cc26c0c0500bfa534dcf02035b76c2d852cea17ad2f3198ab430eaca23", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a0afe723a65013ab6fff82462a9443afdd21e4528133369d689cdfe7ada0030f", "generation_seconds_batch": 20.72816248901654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.905889093875885, "target_z": -0.41406252116743625, "ce": 6.260893821716309, "genuineness": 4.293961524963379, "identity": 0.4662800133228302, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b37e13cc26c0c0500bfa534dcf02035b76c2d852cea17ad2f3198ab430eaca23.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": "b3e2251dbef7a884a9b118c9c484f59d81426d833c002f55a9d35630076c4e51", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5231ce1ebd438ee04fbcb477a207a05be1129269e2199edc8885fec113bbd79d", "generation_seconds_batch": 21.040634370059706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0390994548797607, "target_z": 1.2578125299263754, "ce": 5.708878040313721, "genuineness": 3.7542223930358887, "identity": 0.3492193818092346, "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_RANT_reg/b3e2251dbef7a884a9b118c9c484f59d81426d833c002f55a9d35630076c4e51.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "a162ca4ff093c6cae5b1d1e743f115bdbad7ed5e189933e782b46bb488b0662e", "generation_seconds_batch": 14.689690873026848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4971899390220642, "target_z": -0.7343750102187623, "ce": 6.372091293334961, "genuineness": 4.2751336097717285, "identity": 0.5222494006156921, "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_RANT_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "b4a2e1f07589f66c162ed5e8d7a7d7d0ad8f01494403bc3be4b4e839191874f7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5a94387f985c061417c7f81ef75304ad5f48bcfbfa1989c4babbb0c7287ad921", "generation_seconds_batch": 20.973513899021782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8664660453796387, "target_z": 1.906250014598232, "ce": 5.708878040313721, "genuineness": 3.3211755752563477, "identity": 0.24935339391231537, "hypothesis": "The 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_RANT_reg/b4a2e1f07589f66c162ed5e8d7a7d7d0ad8f01494403bc3be4b4e839191874f7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "b52ff0c83bf4b7337a2a972155b8beb32640dbb08890f8f3f699fc629f7a89c9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ad4a2c9b0c576a14fcc5b492d9fd22d192116521ae1ed3736478aa8f3a269fb", "generation_seconds_batch": 25.156674116966315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.989258050918579, "target_z": 1.2187499970803537, "ce": 5.855817794799805, "genuineness": 3.7291181087493896, "identity": 0.6151221990585327, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b52ff0c83bf4b7337a2a972155b8beb32640dbb08890f8f3f699fc629f7a89c9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "b539dd58ad1d1752211f34cd54bbbbda74db9dcb4bb2a1516b326c533e6528bf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5f70a33a3368e2e7e8948dc2c669d5deeba2096478184cba557931cd17fd43aa", "generation_seconds_batch": 21.18242799595464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7164919376373291, "target_z": -0.5625000058392927, "ce": 6.23706579208374, "genuineness": 4.268857479095459, "identity": 0.702451229095459, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b539dd58ad1d1752211f34cd54bbbbda74db9dcb4bb2a1516b326c533e6528bf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "b56871f2c187ad54cdaa80ed95406e0a188f64524a26896fb92570c4a7a72a61", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0620f05d5725ee52cc7cd2da618e7cefd5780e4291abd93ce9d08332c4ea730a", "generation_seconds_batch": 21.205072573036887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.125641345977783, "target_z": 2.1093751853975453, "ce": 5.091335296630859, "genuineness": 3.5220088958740234, "identity": 0.6484991908073425, "hypothesis": "La puerta sigue abierta! He dejado tu abrigo sobre la silla! ¡Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b56871f2c187ad54cdaa80ed95406e0a188f64524a26896fb92570c4a7a72a61.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "b6141af16335e99611d99290279cca66a8520321a18e4a079ad4ca5cfb7660e9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f45e5b703c2051bb41e1ff982f0a8366016d48f648a349fe9cd65a4e094390cc", "generation_seconds_batch": 22.28690435795579, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.318211078643799, "target_z": 1.476562527006729, "ce": 5.546053409576416, "genuineness": 3.5847692489624023, "identity": 0.6731994152069092, "hypothesis": "La porte 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_RANT_reg/b6141af16335e99611d99290279cca66a8520321a18e4a079ad4ca5cfb7660e9.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": "b627a71b387429ffe4804e796c1de7ae7dc201d92dcc81bd4d227e35535c64f9", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6ecadc54fcb7bfb268dca680050e8390189ebf9c469ea097ae5c298d8f7279f", "generation_seconds_batch": 26.506481830962002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.710146427154541, "target_z": 1.0, "ce": 6.109982967376709, "genuineness": 3.4969046115875244, "identity": 0.5366811752319336, "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_RANT_reg/b627a71b387429ffe4804e796c1de7ae7dc201d92dcc81bd4d227e35535c64f9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "b6da31d7d324aa08afc93b22fa23de137752968ed2ef0a1515a77da10d903461", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "85d6adfca2e8eafdc3e8c7f671e88861925b6c920b966152eea65e5eaf8b29ae", "generation_seconds_batch": 17.173635447979905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.238014817237854, "target_z": -0.9375000408750493, "ce": 5.704906940460205, "genuineness": 4.068024158477783, "identity": 0.4810093641281128, "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_RANT_reg/b6da31d7d324aa08afc93b22fa23de137752968ed2ef0a1515a77da10d903461.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "b7e0eed1c106f0bb5bbf47530ab54a50146ffca424ee1eb401bcf1bf90bc1225", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ed6581d3297a5f43ab6d4ff02df30e390459e836999a4c0475231ee2594925b5", "generation_seconds_batch": 15.703654404962435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8311271071434021, "target_z": -0.47265622700778476, "ce": 6.0146708488464355, "genuineness": 4.231201171875, "identity": 0.5753858089447021, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b7e0eed1c106f0bb5bbf47530ab54a50146ffca424ee1eb401bcf1bf90bc1225.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "b8062b3a534d53a7f13393da4b687318b55a37d36436a0df9e85bae2d7e274f0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6a73551ab631f9873767519e99ddfe2acbac2828c7df38a3233474037a7853f6", "generation_seconds_batch": 22.31954799592495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2031259536743164, "target_z": -0.9648437204385804, "ce": 6.0305562019348145, "genuineness": 4.20609712600708, "identity": 0.6112643480300903, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b8062b3a534d53a7f13393da4b687318b55a37d36436a0df9e85bae2d7e274f0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "b821590553d0ca8d6dfe10428e489c79b08fd38e135d90ebe6c92b2f7355b53d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca39e26c120161b97857c9d12fefc610b97832746a8b6989fc02c2b03a654aca", "generation_seconds_batch": 19.969923511031084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10842728614807129, "target_z": -1.0390625328460217, "ce": 6.0146708488464355, "genuineness": 4.080576419830322, "identity": 0.6508636474609375, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b821590553d0ca8d6dfe10428e489c79b08fd38e135d90ebe6c92b2f7355b53d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "b8f4a62ae8a96ec88246b9310b1e3f5e3601bff1e4dc3a09c8bc8efdf807e306", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f8f33ce19b3b47020fb1351d6658fd7f769c2e2cfdeb970068396e5c4bc04259", "generation_seconds_batch": 22.281048033968545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3401319980621338, "target_z": -0.07373046067535298, "ce": 6.300607204437256, "genuineness": 4.369274139404297, "identity": 0.4589473605155945, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b8f4a62ae8a96ec88246b9310b1e3f5e3601bff1e4dc3a09c8bc8efdf807e306.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "b98e8a1a6a933c4c18cb9005f1804447c4c962d7df7f76eacc40b3e9b7ef9461", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "636c8b07fbc3578c49c61724b91c23e677dc27d3e8769e2755e5d07881cdbe88", "generation_seconds_batch": 16.510164813953452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906339168548584, "target_z": 1.9375000408750493, "ce": 5.77639102935791, "genuineness": 3.427868366241455, "identity": 0.5815770030021667, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b98e8a1a6a933c4c18cb9005f1804447c4c962d7df7f76eacc40b3e9b7ef9461.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "b9bf8f455b20080053f3a154d59dbb51aa777f74b5980ce422a09df75af52430", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f06bfe1bb0cab71bbee84bfd7e090fa9e59294ec42fceaa3acc7f91eab9a59cf", "generation_seconds_batch": 21.52851297101006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3476659059524536, "target_z": -0.8515624686138015, "ce": 6.443574905395508, "genuineness": 4.281409740447998, "identity": 0.6231070756912231, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/b9bf8f455b20080053f3a154d59dbb51aa777f74b5980ce422a09df75af52430.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "ba06a2f7c0c5f58f54c6732b34e872e754d7c0570c43dbcfa778b7467485b3b6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "14efbe7606435221d6c9b2dea395fb7e51d4dd214666dbaa5b9dd126d4c9fa3b", "generation_seconds_batch": 23.11292018101085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.68522572517395, "target_z": 0.9804687335769892, "ce": 6.09409761428833, "genuineness": 3.7793264389038086, "identity": 0.1287785917520523, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ba06a2f7c0c5f58f54c6732b34e872e754d7c0570c43dbcfa778b7467485b3b6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "ba320691832a486609da1eace5e953c0fe153ad48a622d852230a44959e7fc38", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0fa980a0bfde7752d0575ac6dedaef8a21050935527def24c09ba8e389fad59b", "generation_seconds_batch": 21.82428580301348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.1407794952392578, "target_z": -1.2343750102187623, "ce": 6.09409761428833, "genuineness": 4.469690799713135, "identity": 0.4737427234649658, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ba320691832a486609da1eace5e953c0fe153ad48a622d852230a44959e7fc38.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "ba705d97779583e276273609c3177785fd18c7ad040fa912586999443148456a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f942fc7239ab2000f76f9bf3be40a383ce4a426bdd23b2b42331c7cd9b4818d2", "generation_seconds_batch": 17.91086249309592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4709079265594482, "target_z": 0.8125000291964638, "ce": 5.744620323181152, "genuineness": 3.8169827461242676, "identity": 0.7005609273910522, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la cabeza.", "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_RANT_reg/ba705d97779583e276273609c3177785fd18c7ad040fa912586999443148456a.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "db0f4c1dbee77320f7b106155ed42e70f2abf3ab840e8fb33b008b29d5aa705c", "generation_seconds_batch": 14.729028196074069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.1407794952392578, "target_z": -1.2343750102187623, "ce": 6.141753673553467, "genuineness": 4.250029563903809, "identity": 0.6386919617652893, "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_RANT_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "bb60c62d7a40e8fb320d5cc6d24aa098f42a87a3d85aa4c8bff02f19f62f2151", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "24c8fe8047dd8642894b2b7c1143970cca3dd82c9b4bc68b3e20e709bcd54820", "generation_seconds_batch": 24.073972420068458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.537513256072998, "target_z": 1.6484376715292248, "ce": 5.879645347595215, "genuineness": 3.8671910762786865, "identity": 0.16187281906604767, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bb60c62d7a40e8fb320d5cc6d24aa098f42a87a3d85aa4c8bff02f19f62f2151.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": "bb9d0efcf236c15a020ecab16b2a50ab66f83bc15bea517a8857605dc5272e71", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "034a9e016f92e87ab43380d90bcab829ecc9e8803db9f971e694b4a00ce0b1ff", "generation_seconds_batch": 16.955784159014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.681078910827637, "genuineness": 3.6475296020507812, "identity": 0.6973768472671509, "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_RANT_reg/bb9d0efcf236c15a020ecab16b2a50ab66f83bc15bea517a8857605dc5272e71.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "bbad1f3260700d32065c1758f9bd9b67b244b3a8e63a501769d70e4c43c9dd34", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9a441b1f8308e7069abe78ccc9616816ac3bb0f7521c8c9a9dbe2aa97a9c3b3f", "generation_seconds_batch": 23.163964190986007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8098292350769043, "target_z": 1.0781250656920436, "ce": 5.95112943649292, "genuineness": 3.7291181087493896, "identity": 0.3666977286338806, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bbad1f3260700d32065c1758f9bd9b67b244b3a8e63a501769d70e4c43c9dd34.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "bbf4f504a9c824608243f4817e1b0dac5017608b8afdd434861fb2b7d2082f80", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c4e5f17f7bea8562ec4aa288a865edcae97d9744879ea63127e34fbffb832ae", "generation_seconds_batch": 21.449753346969374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8311271071434021, "target_z": -0.47265622700778476, "ce": 6.1020402908325195, "genuineness": 4.137060642242432, "identity": 0.6662917137145996, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bbf4f504a9c824608243f4817e1b0dac5017608b8afdd434861fb2b7d2082f80.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "bc01165e3b3d8adb6e30785943ae1278579955d1100ec43da48ecd9149389c87", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "bb58987f59eb316bfb0498ceb87c2b6dcd5064fab6a36233ac7f82dc20a8f966", "generation_seconds_batch": 19.23910531308502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8977322578430176, "target_z": 0.36328127518195, "ce": 5.911416053771973, "genuineness": 3.961331605911255, "identity": 0.13060887157917023, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bc01165e3b3d8adb6e30785943ae1278579955d1100ec43da48ecd9149389c87.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "bc123b41be9842e6b321370076f029423981e779ee7da4376f7207007b73d11a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "283e49d8c65ac22e8f56b206d49d2e519ddb8422d149b627bdd502fb51347730", "generation_seconds_batch": 16.83859123091679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.524372100830078, "target_z": 2.4218750744509827, "ce": 5.669164657592773, "genuineness": 3.6789097785949707, "identity": 0.34447965025901794, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bc123b41be9842e6b321370076f029423981e779ee7da4376f7207007b73d11a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "bc9da657edf636d9dfe2dc480e6acb0a734dd7388c5190dc50f36d2a892a4525", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "604a16d291516042974c66cb4b23541335663ee37d24dc2a7bf7d11a0eb39578", "generation_seconds_batch": 15.139082290930673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5769361257553101, "target_z": -0.6718750043794696, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.6656261086463928, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bc9da657edf636d9dfe2dc480e6acb0a734dd7388c5190dc50f36d2a892a4525.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "bccaef990d978c5112a17e8af882256b81ae21dcadfc6dc6897cb9d9bb7e3bdd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5eb1e1d709ce956282b102d425ffbbf5260c4d569eb408080c101d89aac87e82", "generation_seconds_batch": 21.52851297101006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.28033530712127686, "target_z": -1.3437500087589391, "ce": 6.4912309646606445, "genuineness": 4.482243061065674, "identity": 0.5599263906478882, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bccaef990d978c5112a17e8af882256b81ae21dcadfc6dc6897cb9d9bb7e3bdd.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "bd311a0226d8512acfb81b9f0789ba99c4c33594b89866eba3ffeac5991fc9c7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a8f710338ee7d44ea6b30841c0276a7c025337fc5aa7fafea8f9aec2615f1d2e", "generation_seconds_batch": 18.83906345802825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0223355293273926, "target_z": 0.46093742043963615, "ce": 5.95112943649292, "genuineness": 3.600459337234497, "identity": 0.3822994828224182, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bd311a0226d8512acfb81b9f0789ba99c4c33594b89866eba3ffeac5991fc9c7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "bd50ef96857342075ab5ae6434a6e869663c0e7f5aab3f0f7c4a75e9a10e08ff", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e071018d000099afb99a52145854515fc8ad12acbfa1d39583a184ac214598cd", "generation_seconds_batch": 20.412216482916847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08849072456359863, "target_z": -1.0546875459844305, "ce": 6.1894097328186035, "genuineness": 4.231201171875, "identity": 0.5811281204223633, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bd50ef96857342075ab5ae6434a6e869663c0e7f5aab3f0f7c4a75e9a10e08ff.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "bda96518edfe6ac8e6e8fdb4c2d7736a89b5782eb13ea4ba616af1c044c3738e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "245a44d6e79052b25592a32c4fc2fb96de453460b60de74c415b1192ff337dbc", "generation_seconds_batch": 23.817347283940762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9693214893341064, "target_z": 1.203124983941945, "ce": 5.80021858215332, "genuineness": 3.7416701316833496, "identity": 0.503517746925354, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bda96518edfe6ac8e6e8fdb4c2d7736a89b5782eb13ea4ba616af1c044c3738e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "bdde5478d820abbac534c5dbe85ddca9154beca3351ce5e6eb01449c9838263b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7aa9c265897802296e591cccdbfa8d322db4c0a4cdb5bf0c437f0d5281a64928", "generation_seconds_batch": 14.996930687921122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5769361257553101, "target_z": -0.6718750043794696, "ce": 6.300607204437256, "genuineness": 4.262581825256348, "identity": 0.5135122537612915, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bdde5478d820abbac534c5dbe85ddca9154beca3351ce5e6eb01449c9838263b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "bde78407cc0543c59808c3d40c071362ea1aef9a5225296e48ba8b7d9ef1e202", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0bb72ab4f47e407ebe362e46db1bf273353d5bad73d0f15b122fec7e15a86ef6", "generation_seconds_batch": 21.456909882021137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7364284992218018, "target_z": -0.546874992700884, "ce": 6.181467056274414, "genuineness": 4.105680465698242, "identity": 0.7387609481811523, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bde78407cc0543c59808c3d40c071362ea1aef9a5225296e48ba8b7d9ef1e202.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "beacfa15822bea0ba287022454345c955bc493ce473dace8c1eeea2526a24dba", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f55272cd6defead384356f36b6a744d438250edece66089b7af2d2bc886006b7", "generation_seconds_batch": 21.789181911968626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4210665225982666, "target_z": 0.773437496350442, "ce": 5.665193557739258, "genuineness": 3.936227321624756, "identity": 0.7934970855712891, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/beacfa15822bea0ba287022454345c955bc493ce473dace8c1eeea2526a24dba.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "bf0592e7e4c716d8dcbc9f6401337abcf6f95102a7d900e278705bac4f4ee618", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1c972662f6ff6c3f69a05817a69ef3a6ef15661c5f5015e1f608ce6d288f18b9", "generation_seconds_batch": 20.404730860027485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.798730194568634, "target_z": -0.4980468733576989, "ce": 6.062326908111572, "genuineness": 4.400654315948486, "identity": 0.7074695825576782, "hypothesis": "La porte 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_RANT_reg/bf0592e7e4c716d8dcbc9f6401337abcf6f95102a7d900e278705bac4f4ee618.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "bf49470ecef84acbc8d223d046c1bfe788f3f480066c8d6401f2c86ffffb6a2d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b779f7f008294c11a20a0fc42276e3e342ad66fe2a630e6005a489990a3fa256", "generation_seconds_batch": 23.153050375985913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.577385902404785, "target_z": 1.6796873240913055, "ce": 5.967014789581299, "genuineness": 3.5345609188079834, "identity": 0.5640272498130798, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bf49470ecef84acbc8d223d046c1bfe788f3f480066c8d6401f2c86ffffb6a2d.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": "bf61e4422e990e1c39875bd151cd82393bb358b75dc10110d2c6247d1f7554cf", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "18043a5b21e63b6f37410c8d63dca06ddeb908c66df34c2c4a2457ab9e75168f", "generation_seconds_batch": 16.240466531948186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4876718521118164, "target_z": 1.6093751386832031, "ce": 5.450741291046143, "genuineness": 3.7040140628814697, "identity": 0.226106196641922, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor! Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bf61e4422e990e1c39875bd151cd82393bb358b75dc10110d2c6247d1f7554cf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "bf6ca0d6bd7b61f04b15555c7ffd2ff4201b447216fd91d7d938ce0f1c3a743b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c00854c01562d8b7feda962b8855baab084e732f70c32001451c939efda2d83d", "generation_seconds_batch": 19.742142562987283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.507608413696289, "target_z": 1.6250001518216117, "ce": 5.824047088623047, "genuineness": 3.6192874908447266, "identity": 0.64893639087677, "hypothesis": "J'ai trouvé la petite boîte ! Elle était près de la vieille fenêtre ! Nous pouvons regarder ensemble ce qu'elle contient !", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bf6ca0d6bd7b61f04b15555c7ffd2ff4201b447216fd91d7d938ce0f1c3a743b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "bfd2fb919a155d6a2f2cd3b2f0f2860a2690728f10b0d8c1d9efab7efa0ff7c5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0b38f139dbc4b452ef3234c6cf23a449c49dec3c3cf3fa2aca2aab1f7239ce18", "generation_seconds_batch": 21.480364601011388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3812510967254639, "target_z": -0.041503917791727095, "ce": 6.141753673553467, "genuineness": 3.9173994064331055, "identity": 0.661625325679779, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bfd2fb919a155d6a2f2cd3b2f0f2860a2690728f10b0d8c1d9efab7efa0ff7c5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "bfe4a6845aacc908b74a0eb4035fe37326f9c9621a48882458906183da863691", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "83004fa274ee6926f17336b477a031e76d7962ff45a98e33942d5ab088b67471", "generation_seconds_batch": 22.99617538100574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6260968446731567, "target_z": 0.15039065802849969, "ce": 6.046441555023193, "genuineness": 3.9676074981689453, "identity": 0.6141373515129089, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bfe4a6845aacc908b74a0eb4035fe37326f9c9621a48882458906183da863691.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "bffb5f48f23d7c435858ba246e71e758782cae0f243a2aeaa34fe03bc5f2fd4a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1fdd3336b9e1bb726dd59e391743ccb8a442e95ab76170c41b57700e4885118f", "generation_seconds_batch": 14.996930687921122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39252305030822754, "target_z": -0.816406282481066, "ce": 6.427689552307129, "genuineness": 4.068024158477783, "identity": 0.5073931813240051, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/bffb5f48f23d7c435858ba246e71e758782cae0f243a2aeaa34fe03bc5f2fd4a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "c0dc846dd3fc19bd5475a77ca77f174e648288de0d080971b03af65ad5db984e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d25e10355a92ac01a6e9c59c1346d3142deab665dadb13b0c0c20903d7ef2291", "generation_seconds_batch": 16.947919137077406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4105329513549805, "target_z": -0.018554656387518263, "ce": 6.109982967376709, "genuineness": 4.080576419830322, "identity": 0.5522714257240295, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c0dc846dd3fc19bd5475a77ca77f174e648288de0d080971b03af65ad5db984e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "c214fb2aac51f57636986144a35787d90840e9d8fa226713fb02fecfefe3ca76", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4adf15f714ba8879bc008d463a77fe046d68cb5ca4697bc6d61fe75916fb4b10", "generation_seconds_batch": 23.992605827050284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8062064051628113, "target_z": -0.49218749343079565, "ce": 5.855817794799805, "genuineness": 4.362998008728027, "identity": 0.5409008264541626, "hypothesis": "La porte 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_RANT_reg/c214fb2aac51f57636986144a35787d90840e9d8fa226713fb02fecfefe3ca76.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": "c24d835cbd35fdcc05d71cb37b552638b3136eb2b4d7c302507fa251eaf51125", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "333d5dc69e8c29096b9592200cf8926ad3d7095d40bf6ca192e8bdbc20182343", "generation_seconds_batch": 30.15688671998214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 5.732706069946289, "genuineness": 3.6914620399475098, "identity": 0.5475895404815674, "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_RANT_reg/c24d835cbd35fdcc05d71cb37b552638b3136eb2b4d7c302507fa251eaf51125.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "c29a534b166b5bb1d3f302027a0ad7caa490946416ebe43a4f69ff012071936c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d4f80125ae1c8f1bef32b0d9c585243b5219d7072667e6b651b5aa3fe24a56d3", "generation_seconds_batch": 22.39996777905617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.852874994277954, "target_z": 0.32812499562053044, "ce": 6.197352409362793, "genuineness": 4.281409740447998, "identity": 0.44638872146606445, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c29a534b166b5bb1d3f302027a0ad7caa490946416ebe43a4f69ff012071936c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "c2d9bbcf43cfa44ae172750b3f4fafb95df5f8a62769de75c12f10771181fa09", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "29fe635d7367f9e01c9d577e20634cb8e150c837deb193b3e9e7ce2c87bc505c", "generation_seconds_batch": 16.510164813953452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9860854148864746, "target_z": 2.0000000934286843, "ce": 5.661221981048584, "genuineness": 3.4466962814331055, "identity": 0.6390203237533569, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c2d9bbcf43cfa44ae172750b3f4fafb95df5f8a62769de75c12f10771181fa09.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": "c2e860f247f30c4f0db83e99718f88edc31a8f22123a61d0d54f3470b8cae444", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3adc44fb2310f757a854b97716f44c75c5a1806ac0aac05a59a2b490e31bea1", "generation_seconds_batch": 23.64320979907643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127002477645874, "target_z": 0.5429687394162819, "ce": 5.998785495758057, "genuineness": 3.6663577556610107, "identity": 0.45328134298324585, "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_RANT_reg/c2e860f247f30c4f0db83e99718f88edc31a8f22123a61d0d54f3470b8cae444.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "c30a7307cd3cfae73194b9f012bbf4e2e1dd8d1330f2f1749a4e61edd88c02f7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b602f7a3fce20035595fbabad7558e9475ad278ed5885e7f5a3250ac537efa3b", "generation_seconds_batch": 21.567104470916092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.663928031921387, "target_z": 2.5312501664198437, "ce": 5.605623245239258, "genuineness": 3.553389072418213, "identity": 0.3347836136817932, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c30a7307cd3cfae73194b9f012bbf4e2e1dd8d1330f2f1749a4e61edd88c02f7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "c34aaafe0675502e6adb0c0f255037445a89476b314a6bab6a6eb96ac3b889b0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ca0703375e62cc1eea499e246451c177a6345550029da8300d86c2d36c1d407f", "generation_seconds_batch": 23.702725843992084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.989938735961914, "target_z": 0.43554686751840616, "ce": 5.959072113037109, "genuineness": 3.6412534713745117, "identity": 0.32849982380867004, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c34aaafe0675502e6adb0c0f255037445a89476b314a6bab6a6eb96ac3b889b0.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "67cae1cce22b6c1b50d834b41cb2af726de479109c09b3fc90d8a482a381fde0", "generation_seconds_batch": 17.343094905023463, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.20058906078338623, "target_z": -1.2812499562053044, "ce": 6.24500846862793, "genuineness": 4.375550270080566, "identity": 0.7236268520355225, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue, sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "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_RANT_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.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": "c3c3a2fd5a638fe555f411d1f229556359e9e3ad2e10df1431719771a3bc1c22", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adc7fd49d95f7cf64e4d1af98cf7775d4e1e729edf100a791bc9433d997f9cca", "generation_seconds_batch": 20.790251467959024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.221701145172119, "target_z": 0.6171875518237232, "ce": 5.855817794799805, "genuineness": 3.7542223930358887, "identity": 0.5312291383743286, "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_RANT_reg/c3c3a2fd5a638fe555f411d1f229556359e9e3ad2e10df1431719771a3bc1c22.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "c43a8f1a7644d00e8b76580a7fbdc44b1681f26badd9a16533629dec105271ba", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "55a4b745824351eb1da0790106b43a660e1c450c1283ccc9765988d9671fcb7d", "generation_seconds_batch": 21.356821218971163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06103324890136719, "target_z": -1.1718749576651275, "ce": 6.268836498260498, "genuineness": 4.3127899169921875, "identity": 0.4999062418937683, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c43a8f1a7644d00e8b76580a7fbdc44b1681f26badd9a16533629dec105271ba.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "c487a4fc1b04458b79fbd354b2a57eae12703b6d0a488bf7e43d065f767a0d1a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90cfbed77df1d8318a846e9526dff689815d5ab71398a7e24db6cefde3d129e5", "generation_seconds_batch": 16.84063144400716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9475736618041992, "target_z": 0.4023438080279718, "ce": 6.197352409362793, "genuineness": 3.923675537109375, "identity": 0.634345531463623, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c487a4fc1b04458b79fbd354b2a57eae12703b6d0a488bf7e43d065f767a0d1a.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "8d22f5b2e4ab9472c90d249c1104beb826518526139c05fbf784ae7e355fe013", "generation_seconds_batch": 17.79267004900612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5470313429832458, "target_z": -0.6953124773727406, "ce": 6.2767791748046875, "genuineness": 4.369274139404297, "identity": 0.7106464505195618, "hypothesis": "Iemand heeft het een 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_RANT_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "c56429739143c310317de20258e352b94173066fc3b41a7877e8169d54740445", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "796fca6ab629cc9567d096c051c6eb5b6b45141de9b8b3b979454bd827c1c794", "generation_seconds_batch": 21.10162761004176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04109680652618408, "target_z": -1.156250037955403, "ce": 5.967014789581299, "genuineness": 4.482243061065674, "identity": 0.6812046766281128, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c56429739143c310317de20258e352b94173066fc3b41a7877e8169d54740445.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "c5c2e5f1b69402f7dd5a147a3735356cb3a1c1ddb22409f7f7655ff2baa27d11", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2b280506f5e48132e0fffbfaa1ede56dabd1dd81262af27a511fc91fbbf46eb6", "generation_seconds_batch": 22.949870598036796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5506539344787598, "target_z": 0.8749998948927303, "ce": 5.990842819213867, "genuineness": 3.8420867919921875, "identity": 0.693793773651123, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c5c2e5f1b69402f7dd5a147a3735356cb3a1c1ddb22409f7f7655ff2baa27d11.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "c5e1bef8f8551f66828749a349e7ffdf74d12ef0e30439ebd2ac16f102925e76", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4aca9ad0ebc0b2b8e0b92a1e2fde3ef837efb45998c0b77ce7754115f65f566b", "generation_seconds_batch": 25.31258234498091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.100270390510559, "target_z": -0.2617187832109776, "ce": 5.808161735534668, "genuineness": 4.105680465698242, "identity": 0.6306889653205872, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c5e1bef8f8551f66828749a349e7ffdf74d12ef0e30439ebd2ac16f102925e76.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "c5fabb17bab4e191aa2f1ead6edaacd671ceebc1073743336a917d18596b2c36", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "29653729a5715f1f3646ce01a60ea373960a1ef6e016192f0139f07a32b9561b", "generation_seconds_batch": 23.563897556974553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.161891460418701, "target_z": 0.5703125124084971, "ce": 6.141753673553467, "genuineness": 4.18726921081543, "identity": 0.5535547137260437, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c5fabb17bab4e191aa2f1ead6edaacd671ceebc1073743336a917d18596b2c36.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "c6219184103669a6c5bc58bc3e378466a3478a72c9c94d5d613a1f6b5518c650", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a9c2b6fc99fd7662ede3261d6bc85a5f28a4c1ce94ec07eaf1f60e781c8e8844", "generation_seconds_batch": 26.001607366953976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.045895099639893, "target_z": 2.04687513284391, "ce": 5.780362129211426, "genuineness": 3.565941095352173, "identity": 0.529178261756897, "hypothesis": "La porte 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_RANT_reg/c6219184103669a6c5bc58bc3e378466a3478a72c9c94d5d613a1f6b5518c650.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "c688da5935047186cf5430f2f18da7183595e3d9ebb64b8bf0104bf532b0dc8b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4843a9dbc3456d4046d8533f0dccf40137d87ce20416fad62d42d28683cccc90", "generation_seconds_batch": 21.56533033098094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.631080985069275, "target_z": 0.15429691131310186, "ce": 5.816103935241699, "genuineness": 3.8671910762786865, "identity": 0.7564395070075989, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c688da5935047186cf5430f2f18da7183595e3d9ebb64b8bf0104bf532b0dc8b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "c6fb71dbafe4bb01319de2d07e46f672880d874edd8367da9a195fa71c2d04bb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d616545afd48cc4aecb68df91134408dcaa2776d7520af51f051188c01db8642", "generation_seconds_batch": 25.269430187996477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.228496551513672, "target_z": 1.4062499678838898, "ce": 5.847874641418457, "genuineness": 3.6067352294921875, "identity": 0.3828977942466736, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c6fb71dbafe4bb01319de2d07e46f672880d874edd8367da9a195fa71c2d04bb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "c71826be4ae3770496155b64ab11db2d6bf5a9e7f9e76643e9bd5d24f15411d8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "97258560472767de9645e4f3783397ded95cba958ddc8072c851d12a59d5ae6a", "generation_seconds_batch": 16.064549545990303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0690040588378906, "target_z": 1.2812498627766202, "ce": 5.696964263916016, "genuineness": 3.587907314300537, "identity": 0.34139934182167053, "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_RANT_reg/c71826be4ae3770496155b64ab11db2d6bf5a9e7f9e76643e9bd5d24f15411d8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "c7701f183770032119cfefba7d39e85855023a934d6d86f5da4a9fa9a37920ff", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ae64098682cc534f0c78370d4f88697dcaaf3e1d1ded35041ab39c6820736d0b", "generation_seconds_batch": 17.44326017901767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2778879404067993, "target_z": -0.9062500145982318, "ce": 6.284721851348877, "genuineness": 4.074300289154053, "identity": 0.5035792589187622, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c7701f183770032119cfefba7d39e85855023a934d6d86f5da4a9fa9a37920ff.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "c79776fc4b7b749156926c0e7cb8bd6cbf2bc6debb0728151c8eccce84207430", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "5c816bd46dd1c09a1d91c9509a11a1efa20c1ea90154c7eab66163a4db2841e8", "generation_seconds_batch": 20.7179680459667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2722231149673462, "target_z": -0.12695313832088662, "ce": 6.157639026641846, "genuineness": 3.835810661315918, "identity": 0.24930307269096375, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c79776fc4b7b749156926c0e7cb8bd6cbf2bc6debb0728151c8eccce84207430.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "c7b5e0c0db292f7bab1ca643d5f127e97c442ddccc8efbd268e424e4cc463147", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "76dd62ac6c8ef9ea7f4403048cf9c14ec4fdb82a0468fa1dacee6db491baa9d8", "generation_seconds_batch": 23.082436795928515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.085768222808838, "target_z": 2.0781251591207277, "ce": 6.117925643920898, "genuineness": 3.7040140628814697, "identity": 0.2644439935684204, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c7b5e0c0db292f7bab1ca643d5f127e97c442ddccc8efbd268e424e4cc463147.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "c804aeae980a18a79fcc6b2eb6ee9880e0c9661d6bde23769f276d6f36e3db21", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6fb10776898ebf88e21636be970bba38a5f1567583df3faf34b866533c3a8a6", "generation_seconds_batch": 16.791696242056787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8265929222106934, "target_z": 1.8749999883214146, "ce": 5.982900142669678, "genuineness": 3.418454170227051, "identity": 0.5980220437049866, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c804aeae980a18a79fcc6b2eb6ee9880e0c9661d6bde23769f276d6f36e3db21.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "09532502a357ee3e85e14159ea2d56c016e4d4a4e80a1b180d3494cd73334d8f", "generation_seconds_batch": 17.457241611904465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.49220579862594604, "target_z": -0.7382812635033645, "ce": 6.23706579208374, "genuineness": 4.457139015197754, "identity": 0.7644360661506653, "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_RANT_reg/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "c82238dcacf63084d153515b8f97210f1da4732d1358450ca46b95f3fc04981c", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4adf4f61490961d6c9d14f54ea7d2d4c15c501b53c85c4675d5f88cd52e26a9", "generation_seconds_batch": 21.456909882021137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3975071907043457, "target_z": -0.8125000291964638, "ce": 6.1894097328186035, "genuineness": 4.362998008728027, "identity": 0.7140395641326904, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c82238dcacf63084d153515b8f97210f1da4732d1358450ca46b95f3fc04981c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "c83e4f80bea717639401f947d8e174935d1aa12371d9377e889f807bb6f3efa6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7c2b9b865c28b4be1702a0094da953d5b039434aac6ce7ed0b9b35a7d3438b70", "generation_seconds_batch": 17.087409920990467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.525733470916748, "target_z": 0.8554688153270877, "ce": 5.824047088623047, "genuineness": 3.625563621520996, "identity": 0.32231852412223816, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c83e4f80bea717639401f947d8e174935d1aa12371d9377e889f807bb6f3efa6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "c86e6b0c3082d836cf4ffd358e8562796c7af47973e9e5bbf8edf86efccc1400", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "682575e852c558dc24779599624cc41fec24cdbbb627c94b2c1b207a45c417aa", "generation_seconds_batch": 26.610232162987813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.620431900024414, "target_z": 0.9296874408771608, "ce": 5.839932441711426, "genuineness": 3.6349775791168213, "identity": 0.5549970865249634, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c86e6b0c3082d836cf4ffd358e8562796c7af47973e9e5bbf8edf86efccc1400.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "c88c6fd5cd76fa1e9155f7039cab9881a9ed3e92a032655c5ec1dd0e2cf304bf", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ad632d33a8d2c9df1f5f9def2ea9c771e6636ad1395446096b5835eb0868fa97", "generation_seconds_batch": 17.44326017901767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13833212852478027, "target_z": -1.0156250131384088, "ce": 6.316492557525635, "genuineness": 4.21237325668335, "identity": 0.47405344247817993, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c88c6fd5cd76fa1e9155f7039cab9881a9ed3e92a032655c5ec1dd0e2cf304bf.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "c0e62144be46feb5d99ff81668990052459053b31f3a538a9454dc9a3ab829d9", "generation_seconds_batch": 15.66625829495024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.297824501991272, "target_z": -0.8906250014598232, "ce": 6.38003396987915, "genuineness": 4.482243061065674, "identity": 0.5278474688529968, "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_RANT_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.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": "c95bdb9d0ab15bbd66b1beaab17cd2132b37e4598c496ecf17cf8378e61bc2ea", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40f70faa4dc4b76237c6d65bc0d39d539be43e5dc6deafc3d4a9a04402529b8b", "generation_seconds_batch": 29.958361276076175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.258401393890381, "target_z": 1.4296874875915029, "ce": 5.95112943649292, "genuineness": 3.7667744159698486, "identity": 0.6813883781433105, "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_RANT_reg/c95bdb9d0ab15bbd66b1beaab17cd2132b37e4598c496ecf17cf8378e61bc2ea.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "c97fbc013e2b2cecce1fb0602faaca33c4d61ed6072eb01a0ee8e1dcdea303f6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e4e53652976093f1afb21307dae3d3072b44ae837ae83008246ba849f66e58b0", "generation_seconds_batch": 15.238079856033437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.584977626800537, "target_z": 0.11816402171618963, "ce": 5.49045467376709, "genuineness": 3.9425034523010254, "identity": 0.6314638257026672, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c97fbc013e2b2cecce1fb0602faaca33c4d61ed6072eb01a0ee8e1dcdea303f6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "c9dd63601fe5223706b0a277e607131aea223eaaa6632cb91b4cee5ac423cf89", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "afc53100e9460f53c14fc6a94d10b7b1c7e5e72c09dfa241206042c3a4abe0d2", "generation_seconds_batch": 25.801488447003067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2484331130981445, "target_z": 1.4218749810222986, "ce": 6.086154937744141, "genuineness": 3.7793264389038086, "identity": 0.24384677410125732, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/c9dd63601fe5223706b0a277e607131aea223eaaa6632cb91b4cee5ac423cf89.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "ca14007c760bd5fdeeaa41e7818bb94ac12d5b7799708964d1f1ef79f3d0a9aa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cb96b6a8b89cbe3eab90e02588de88af91389413517d71eadbeb8c060a3bef0d", "generation_seconds_batch": 19.756685576052405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07852256298065186, "target_z": -1.0624999591249507, "ce": 6.3641486167907715, "genuineness": 4.2249250411987305, "identity": 0.4657418131828308, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ca14007c760bd5fdeeaa41e7818bb94ac12d5b7799708964d1f1ef79f3d0a9aa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "cb27bc800aceae99fd8085d6f1622ff04292dcb565b12cfb0799e90da54485ac", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "9407bc954515ee9b30cbfb16c83bdcd36d34e0b5910ecc697d5ffe6b2680814c", "generation_seconds_batch": 20.01793949108105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6603050231933594, "target_z": 0.9609374671539782, "ce": 5.80021858215332, "genuineness": 3.911123275756836, "identity": 0.28703513741493225, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cb27bc800aceae99fd8085d6f1622ff04292dcb565b12cfb0799e90da54485ac.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "cb4bf8785b3412a03261b3a500d271e903700fa07e24051ec9a1c3a43bbf334e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "763dbef281e1889b50c3efdf8e0e6d92d6e7909320e550f1ff694288057b8cb5", "generation_seconds_batch": 20.7179680459667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.987446665763855, "target_z": 0.4335937408761051, "ce": 5.990842819213867, "genuineness": 3.613011360168457, "identity": 0.439960777759552, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cb4bf8785b3412a03261b3a500d271e903700fa07e24051ec9a1c3a43bbf334e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "cc0448c99d1aad7ad0ceaa0de4a0030d5a28b078ea0ed14ab61d91b8a597b60a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbb6701024d8c6b5ca19fa9e6fcdd859da211c50eda587bbb2595e652a3f5edf", "generation_seconds_batch": 16.74751420796383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.211732864379883, "target_z": 0.6093750452545189, "ce": 6.022613525390625, "genuineness": 3.7793264389038086, "identity": 0.444689005613327, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cc0448c99d1aad7ad0ceaa0de4a0030d5a28b078ea0ed14ab61d91b8a597b60a.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": "cc4de614b58fa04b4b6858b4225e44321c155864e83527384d80c417e875fdd9", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a04d032fbca61a1fb96934b0c554d67b4621abf963c1166ed413d5ce963ba62", "generation_seconds_batch": 21.663625792018138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680241584777832, "target_z": 0.9765624802923869, "ce": 6.109982967376709, "genuineness": 3.4341442584991455, "identity": 0.5356724858283997, "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_RANT_reg/cc4de614b58fa04b4b6858b4225e44321c155864e83527384d80c417e875fdd9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "cc85cd3df721796855845dd7f31ba489892db7cfa680ae338c325b1c4b4a8b92", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c39fb1ef54be0000dbf5a36d08960a850dea005ce244f6588192b73338b754a8", "generation_seconds_batch": 21.08569036202971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47725343704223633, "target_z": -0.7499999766428289, "ce": 5.982900142669678, "genuineness": 4.231201171875, "identity": 0.6238178610801697, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cc85cd3df721796855845dd7f31ba489892db7cfa680ae338c325b1c4b4a8b92.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "cc8f21be5569e26f64de83a843e981a75a8dc941412baa4226fa31f8df0936d2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f212afb0658523eac5b687f1e595f165a6b5b4177b987c0b90a327fa91e2b657", "generation_seconds_batch": 16.94173896301072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7169418334960938, "target_z": 1.7890624160601665, "ce": 5.982900142669678, "genuineness": 3.565941095352173, "identity": 0.07862623035907745, "hypothesis": "La porte 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_RANT_reg/cc8f21be5569e26f64de83a843e981a75a8dc941412baa4226fa31f8df0936d2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "cd620166a8db57f43a337ad3251af58ae1a86d9581149753601648d62d65c34b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c68eae4ddbe99dece06472b7f185a43798aea804b46dd1194daf45d4c7acb01", "generation_seconds_batch": 25.471652873093262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.577385902404785, "target_z": 1.6796873240913055, "ce": 5.299830436706543, "genuineness": 3.6851859092712402, "identity": 0.6480686068534851, "hypothesis": "La puerta sigue abierta! He dejado tu abrigo sobre la silla! Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cd620166a8db57f43a337ad3251af58ae1a86d9581149753601648d62d65c34b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "ce0a9071c02d9bb206320b7f1db808fba6afe6a982380ebf94b363b3a9f1e32d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6ce3204c1a1c773399b1602f65cf436534f0a792491b24d79e9dbdb584bcbb85", "generation_seconds_batch": 22.071215928997844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0572245121002197, "target_z": 0.4882811934318514, "ce": 6.141753673553467, "genuineness": 3.9299514293670654, "identity": 0.7027503848075867, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ce0a9071c02d9bb206320b7f1db808fba6afe6a982380ebf94b363b3a9f1e32d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "ce8d619463787364c322c074b08d03eb06ab60d45d0a4f5af3e41d7604a560ef", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d3a4d4bf3ec0638d4ca4615a243438d5d7d2248e5bfe1259d553757a135e60b8", "generation_seconds_batch": 17.646514503052458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3849892616271973, "target_z": -0.03857418111393338, "ce": 5.919358730316162, "genuineness": 4.068024158477783, "identity": 0.7271430492401123, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ce8d619463787364c322c074b08d03eb06ab60d45d0a4f5af3e41d7604a560ef.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": "ceb52bac3832dd8bb98361abdd1dc830cf5c7ce72300866560eddf443f43ffe8", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95de4d69655fd68dd2bf4f64c7e28b0094dfcf2b27bc8ab9d6f1bf85e4c8ba24", "generation_seconds_batch": 21.62205734103918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8696389198303223, "target_z": 1.1250001051072698, "ce": 6.0146708488464355, "genuineness": 3.591045379638672, "identity": 0.4360727071762085, "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_RANT_reg/ceb52bac3832dd8bb98361abdd1dc830cf5c7ce72300866560eddf443f43ffe8.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": "ced808824bc7a4f2227ef7177806cd5778232380e541c40ed5c50a3a93bc2ff4", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4044f2abe2bad6d8bd304949904ff1d73b8b43c9beba57aa7ef426f8a6a98ca8", "generation_seconds_batch": 21.137362504028715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.371225118637085, "target_z": 0.7343749635044202, "ce": 5.855817794799805, "genuineness": 3.785602569580078, "identity": 0.6179100275039673, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ced808824bc7a4f2227ef7177806cd5778232380e541c40ed5c50a3a93bc2ff4.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "cf602f374e8945cb999a0e281891eafed3e7215ba0eaca52c2f6b6d6728650bd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9569cdae99fefd78782905cbe9503989996a57b0df7f88f0454d31fdaadadc1", "generation_seconds_batch": 20.80338961700909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.667100429534912, "target_z": 1.7499998832141448, "ce": 6.125868320465088, "genuineness": 3.6914620399475098, "identity": 0.5032495260238647, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cf602f374e8945cb999a0e281891eafed3e7215ba0eaca52c2f6b6d6728650bd.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "5e35222945d11971eb4e7a4fdf8fae8eb9d473f5da2be020a2916b07218fd504", "generation_seconds_batch": 15.248356981086545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20811009407043457, "target_z": -0.9609374671539782, "ce": 6.316492557525635, "genuineness": 4.356721878051758, "identity": 0.5022384524345398, "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_RANT_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "cf70f8c7fd797f169c51e0c146904680d78fa94809677673fc7a2767d8a91882", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "29660d0958928a29d22bdd2b06b41e0c6f53cbf0a8c95fbc24594b42604d1de8", "generation_seconds_batch": 15.836484865052626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.1407794952392578, "target_z": -1.2343750102187623, "ce": 6.086154937744141, "genuineness": 4.218649387359619, "identity": 0.686626672744751, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cf70f8c7fd797f169c51e0c146904680d78fa94809677673fc7a2767d8a91882.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "cfdde604c52f120570af3db83f66b74a90a6784339601e8c59a53cfc20823297", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "0dd96e2d00409abdafcbf4381e19d4daa6f7cf23fa67e713db067c2d970bb07e", "generation_seconds_batch": 21.0751173410099, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.02595853805542, "target_z": 2.0312501197055015, "ce": 5.11317777633667, "genuineness": 3.5220088958740234, "identity": 0.5785653591156006, "hypothesis": "¡La puerta sigue abierta! ¡He dejado tu abrigo sobre la silla! ¡Dime cuándo estás listo?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cfdde604c52f120570af3db83f66b74a90a6784339601e8c59a53cfc20823297.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "cfe58d97925a833538a52590e2068d4f508dcd311ba5d370df1ab189eff8c97b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fb9f26c23371acfb6c1bf220c689207823199aab746148ebcc082fa7cbda66f7", "generation_seconds_batch": 22.950990261044353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8585398197174072, "target_z": -0.45117188065681485, "ce": 6.197352409362793, "genuineness": 4.011539936065674, "identity": 0.6226807236671448, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cfe58d97925a833538a52590e2068d4f508dcd311ba5d370df1ab189eff8c97b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "cffd2822604b71a4a87d9f8778fe58699c6fc6186576ae99fbc8b9a5e82eb462", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "fb693049c0df1f5dbb35c4d9bccd3d63dc2b63519acfeee83f3228309fffc658", "generation_seconds_batch": 23.554372834041715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8995437622070312, "target_z": 1.1484376248148827, "ce": 5.796247482299805, "genuineness": 3.6349775791168213, "identity": 0.5275808572769165, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/cffd2822604b71a4a87d9f8778fe58699c6fc6186576ae99fbc8b9a5e82eb462.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": "d08997ad9a2ad5596abb8ad52c1f9e9c59686681eb10deb398aaa4368e37bae6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35ae207d81ba8b158ac3916ef9ecdf0c3a165453f2fc9dc234de1833bf2f082d", "generation_seconds_batch": 29.958361276076175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4709079265594482, "target_z": 0.8125000291964638, "ce": 5.732706069946289, "genuineness": 3.8169827461242676, "identity": 0.659754753112793, "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_RANT_reg/d08997ad9a2ad5596abb8ad52c1f9e9c59686681eb10deb398aaa4368e37bae6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "d10ad99dab126c320a816195f87a2b4169b72dd28d83a3ca73ba06666f2321ae", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b97fb15fb8ce270ccc0d1e88b655c019fa83dcdfbb18d17ed42faed777f91b0c", "generation_seconds_batch": 18.295853212941438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1886234283447266, "target_z": 1.3749999416070724, "ce": 5.391171455383301, "genuineness": 3.71028995513916, "identity": 0.6658483743667603, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d10ad99dab126c320a816195f87a2b4169b72dd28d83a3ca73ba06666f2321ae.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "f2de5843592e0ffdb1211019919777fad6c1e1be48b94084da5598e2be37fc14", "generation_seconds_batch": 15.08116825891193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0585860013961792, "target_z": -1.0781249722633595, "ce": 6.372091293334961, "genuineness": 4.325342178344727, "identity": 0.533252477645874, "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_RANT_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "d169bb1eb1f5705a147e7604f033a31b378b8b73f1ade40f13ba318eb96bccfe", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ee81bea6401b8513810d4496a2e94e5bc115a61895c127f055dd41bb52118946", "generation_seconds_batch": 24.958994923043065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9860854148864746, "target_z": 2.0000000934286843, "ce": 5.911416053771973, "genuineness": 3.7667744159698486, "identity": 0.3766237497329712, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d169bb1eb1f5705a147e7604f033a31b378b8b73f1ade40f13ba318eb96bccfe.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "d1c1c4bdf632847462024d6834a9efce7202547c9676b08b5a0c4507910fd8fa", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a521eeca798323f7a731c05f74c2da6b66411e43785d593dcf32725cad11c393", "generation_seconds_batch": 23.817347283940762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.507608413696289, "target_z": 1.6250001518216117, "ce": 6.006728172302246, "genuineness": 3.6663577556610107, "identity": 0.3981834053993225, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Tönster. 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_RANT_reg/d1c1c4bdf632847462024d6834a9efce7202547c9676b08b5a0c4507910fd8fa.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "d1e0a65bf5534b03d0c8c976d6cd96492a2e2c14bd86a97e4f60874cca1c6a9f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "586c245b3f7bbc1062e37b1b5a5a27f54845aaf0c8fa94419914de37d77041d8", "generation_seconds_batch": 19.885150048998185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34268176555633545, "target_z": -0.8554687218984036, "ce": 6.316492557525635, "genuineness": 4.15588903427124, "identity": 0.4410761594772339, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d1e0a65bf5534b03d0c8c976d6cd96492a2e2c14bd86a97e4f60874cca1c6a9f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "d1f0d044ee3aa9869455d9324865f97837d71c94a036abd540e3c523c7ff5282", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "fb1f1bd5482a3160bb1e02fb9137b04af34da231709f4872e065dc03859141d5", "generation_seconds_batch": 21.789181911968626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9151766300201416, "target_z": 0.37695306824937347, "ce": 5.8875885009765625, "genuineness": 4.042920112609863, "identity": 0.6578922271728516, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d1f0d044ee3aa9869455d9324865f97837d71c94a036abd540e3c523c7ff5282.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": "d210698a035a8900ea878cf714073cb3064114128df04ffabcb22c3acf0e35af", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ac498e0cc98f8ecab7239bfc685c67477b0cf0eb577bd542d60ddfde57c6f4d", "generation_seconds_batch": 24.671221949975006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9693214893341064, "target_z": 1.203124983941945, "ce": 5.95112943649292, "genuineness": 3.8671910762786865, "identity": 0.6399102210998535, "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_RANT_reg/d210698a035a8900ea878cf714073cb3064114128df04ffabcb22c3acf0e35af.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "d2238628ccb0f75836bb385b9be5aaf3c6dceca023ed6542e38a104005bd66c2", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b367b2a4482708d740631757c7e9437595e1af229bb552d6bfe29943f46caebb", "generation_seconds_batch": 24.296484290040098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9693214893341064, "target_z": 1.203124983941945, "ce": 5.974957466125488, "genuineness": 3.7793264389038086, "identity": 0.289287269115448, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d2238628ccb0f75836bb385b9be5aaf3c6dceca023ed6542e38a104005bd66c2.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "4c3055f6304686b669c10861ab752269d7076bf2174e499443d2c206e4f41377", "generation_seconds_batch": 17.694810020038858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21309423446655273, "target_z": -0.9570312138693761, "ce": 6.181467056274414, "genuineness": 4.469690799713135, "identity": 0.6755927801132202, "hypothesis": "Las llaves están en la mesa de la cocina, mire en el cajón, y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "d232cdd90835544a98442b3886feb799355dd0fea953d66df4519ea21c71a18a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "0c2f9be9d09cb7932e48f4c2f38e031152f287e76f81e69d4bfa9cc242cd0693", "generation_seconds_batch": 23.25064266996924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.495828628540039, "target_z": 0.8320312956194746, "ce": 5.538110733032227, "genuineness": 3.798154592514038, "identity": 0.6930145621299744, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "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_RANT_reg/d232cdd90835544a98442b3886feb799355dd0fea953d66df4519ea21c71a18a.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": "d28367a88c963b76b7744edbdafbe37952e50aaec40d6ad5be84827b1c21dc89", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c22af7774b9ce574f41cfee610ad0268cb2f38c60637278f3784f9200c1f826", "generation_seconds_batch": 16.18593834794592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.756814956665039, "target_z": 1.820312442336984, "ce": 5.593709468841553, "genuineness": 3.6287014484405518, "identity": 0.6702967286109924, "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_RANT_reg/d28367a88c963b76b7744edbdafbe37952e50aaec40d6ad5be84827b1c21dc89.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "d285317590c80e7e1a2f158d2b72bad4c373bed01fb43a5e1c7d0a17f8c80590", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "fbfbf52f168929440c7681dd47c466718b56994a4d98d52644dcfa98ee862ab7", "generation_seconds_batch": 20.445083676022477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18817353248596191, "target_z": -0.9765624802923869, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.631726861000061, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d285317590c80e7e1a2f158d2b72bad4c373bed01fb43a5e1c7d0a17f8c80590.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "d2a7253efd8fc07589d86d96c5338e5253f6fc13581aec0fc3e9e1e0f2d84d25", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c9f8b114f2e0d8eaf12287debebcef304d2c5aa1fafb36866997dbce3b509863", "generation_seconds_batch": 25.029188922024332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.447798728942871, "target_z": 1.5781251124063855, "ce": 5.919358730316162, "genuineness": 3.4812145233154297, "identity": 0.48634034395217896, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d2a7253efd8fc07589d86d96c5338e5253f6fc13581aec0fc3e9e1e0f2d84d25.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "d2fff103fabbf47a4d70a8138342dc67e6df64c398ebeec836136278217e3ff8", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "402141cd63927d1651462716b31539e05ae8ce51a264c157fe7691b829462e1b", "generation_seconds_batch": 22.36113321292214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7164919376373291, "target_z": -0.5625000058392927, "ce": 6.1894097328186035, "genuineness": 4.13078498840332, "identity": 0.5776405334472656, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d2fff103fabbf47a4d70a8138342dc67e6df64c398ebeec836136278217e3ff8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "d3016a371a9cdc430b2a1c32a7aa7d6fa8b8943eba8f92731f8cbb3ebe292ae7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d6275bcb2cf34123121ca9856993016a23de2a800b459e30435dfaa79bdbcb38", "generation_seconds_batch": 22.71839511510916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.912684679031372, "target_z": 0.3750000350357566, "ce": 6.23706579208374, "genuineness": 4.256305694580078, "identity": 0.652283251285553, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d3016a371a9cdc430b2a1c32a7aa7d6fa8b8943eba8f92731f8cbb3ebe292ae7.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": "d388e06d63827dd1c951fbe4df99865efefb09f9c7dbcdbf6af58d04c54f610c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "9b54fdfdac6bb43c02769c2ae0b4f7525330c4776bee171ed246ddad68836a3a", "generation_seconds_batch": 16.724314004997723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.006021976470947, "target_z": 2.015625106567093, "ce": 5.967014789581299, "genuineness": 3.7793264389038086, "identity": 0.462790846824646, "hypothesis": "Eine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können!", "wer": 0.041666666666666664, "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_RANT_reg/d388e06d63827dd1c951fbe4df99865efefb09f9c7dbcdbf6af58d04c54f610c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "d3a87a0eeb809f48dae13d2ea5fe75782466abaac811202bade157493d1dc7fd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "231f8ae525e0ad00c4a8b06f030fabb4e1f036bfbe7c084a004b0e58578cd6aa", "generation_seconds_batch": 19.914919547969475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7169418334960938, "target_z": 1.7890624160601665, "ce": 5.39911413192749, "genuineness": 3.6287014484405518, "identity": 0.2681855857372284, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d3a87a0eeb809f48dae13d2ea5fe75782466abaac811202bade157493d1dc7fd.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "ed9e4d8490b86bc8dc4ea5be795a4e95e69e5b4ac1cbf4ecba385692b56f90ed", "generation_seconds_batch": 17.845271227066405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04109680652618408, "target_z": -1.156250037955403, "ce": 5.935244083404541, "genuineness": 4.469690799713135, "identity": 0.7293071150779724, "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_RANT_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "d45b951e31515d9d33c500129767495577257ad350f46edbe0a56d417e071648", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "a26a02caf58a9bb2c86e87e54984778c0b95c2e1193e4e5e5cd087b7c306fd9c", "generation_seconds_batch": 21.406599993933924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5044355392456055, "target_z": 2.406250061312574, "ce": 5.33954381942749, "genuineness": 3.4498343467712402, "identity": 0.5134819149971008, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt! Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d45b951e31515d9d33c500129767495577257ad350f46edbe0a56d417e071648.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "d47ff715e465e6f702af7cb3feb2962ee59d4b3449702fe9b1771159bb6c7ef0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2e885adb61f8db42d68afde809e382aec40d1d90ade3c5ffed155f5bdd38177b", "generation_seconds_batch": 23.908043537056074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.436018943786621, "target_z": 0.7851562562042486, "ce": 5.895530700683594, "genuineness": 3.8420867919921875, "identity": 0.6434199213981628, "hypothesis": "En he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d47ff715e465e6f702af7cb3feb2962ee59d4b3449702fe9b1771159bb6c7ef0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "d543bcfce0208e093f12fb6ed502bee992c210488dcf1e1f8a0d4fa7d58ce99b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57894f09c7def5a00bbef9c719e37bd598eec5febca6e432c4643e1c52fe1478", "generation_seconds_batch": 16.458508905023336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23303067684173584, "target_z": -0.9414062941596515, "ce": 6.109982967376709, "genuineness": 4.2751336097717285, "identity": 0.6177672147750854, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d543bcfce0208e093f12fb6ed502bee992c210488dcf1e1f8a0d4fa7d58ce99b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "d5460a26de6ed44136283e6dcddb1c068777f4330ec7880f477dadaf26b820c0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3c478f2ba9ecd468d6710082e6a5813a0045af7cda743e95991ffff4cd1cf955", "generation_seconds_batch": 21.641978056984954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3825547695159912, "target_z": -0.8242187890502704, "ce": 6.229123115539551, "genuineness": 4.105680465698242, "identity": 0.7746305465698242, "hypothesis": "La porte 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_RANT_reg/d5460a26de6ed44136283e6dcddb1c068777f4330ec7880f477dadaf26b820c0.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": "d578a33c12b76832ef4889151b51ddc2511ae9c83553ad6cadf89820081362c2", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3f82b84551894e12264480c1409dee2d3744429dcde50e3c9086677f8ac294c", "generation_seconds_batch": 24.592902717995457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.378020763397217, "target_z": 1.523437566421955, "ce": 5.990842819213867, "genuineness": 3.6067352294921875, "identity": 0.31478023529052734, "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_RANT_reg/d578a33c12b76832ef4889151b51ddc2511ae9c83553ad6cadf89820081362c2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "d57d18c0f0c856f2dce2c2c279ff304c9c7b8d8b25a8fadf506ead4c102776f1", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c9f5a1eed3eae5a98b5231f2fda744b941ff0c9da4325aed6ed520a080d96e8d", "generation_seconds_batch": 20.855221818899736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5157651901245117, "target_z": 0.8476563087578834, "ce": 5.391171455383301, "genuineness": 3.71028995513916, "identity": 0.33269166946411133, "hypothesis": "I found the little box! It was beside the old window! We can look inside together!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d57d18c0f0c856f2dce2c2c279ff304c9c7b8d8b25a8fadf506ead4c102776f1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "d5a2364e9fc3b0b84a78bba4b6f4aeb5a688c5cefd658c359ab86767b249a68a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5a2d15cce8481d9df58b8ef78317a07f9da9b742ad96c2e96738967dc0ff61ec", "generation_seconds_batch": 24.352225829963572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4553900957107544, "target_z": 0.016601529745217176, "ce": 6.09409761428833, "genuineness": 3.8295347690582275, "identity": 0.6182214617729187, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d5a2364e9fc3b0b84a78bba4b6f4aeb5a688c5cefd658c359ab86767b249a68a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "d61e52305df17593ea3c2a765e91823cf7122efd1c3ba1e2a9dc425ed2e3d449", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7432d483b6e947f9bda61329000a57aa25139bacda4fef9c005ba653a48b9037", "generation_seconds_batch": 23.334325533010997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5175766944885254, "target_z": 1.632812658390816, "ce": 5.80021858215332, "genuineness": 3.8169827461242676, "identity": 0.23123329877853394, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d61e52305df17593ea3c2a765e91823cf7122efd1c3ba1e2a9dc425ed2e3d449.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "d647a880b503fccadd13b56cb7060f15b1107334087d416b3ca2b1951e176acf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af4ebba600771359ef51986d55c4c2a316213afda1315321e16767cf364c976f", "generation_seconds_batch": 23.354401162010618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.311415672302246, "target_z": 0.6875001109465624, "ce": 6.197352409362793, "genuineness": 3.848362922668457, "identity": 0.5789438486099243, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d647a880b503fccadd13b56cb7060f15b1107334087d416b3ca2b1951e176acf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "d6a80af81b1ee3f25abe3acd137ce079db82bcec78bbf288d454ac28cde2ba10", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "0eaa7275672e0330ca6721c92bf132c8e76c1d82f6d24147fc9bc34bbb4b12a4", "generation_seconds_batch": 25.881722176913172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7001781463623047, "target_z": 0.9921874934307956, "ce": 5.847874641418457, "genuineness": 3.500042676925659, "identity": 0.35112133622169495, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d6a80af81b1ee3f25abe3acd137ce079db82bcec78bbf288d454ac28cde2ba10.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "d6ad35ae6130a87e05fa6cd95fac382190a2cadec9eeec9b49bf770b62a0b71e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0e7bc714eb33feff3dbfee0d3cf136e0276a83188b98459f945ebb47cfff266", "generation_seconds_batch": 20.11865293898154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1949690580368042, "target_z": -0.1874999708035362, "ce": 6.205295085906982, "genuineness": 4.111956596374512, "identity": 0.6561868786811829, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d6ad35ae6130a87e05fa6cd95fac382190a2cadec9eeec9b49bf770b62a0b71e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "d6ef590efe590f32f6d58c22a4d0f47a47011de7e5ded8d65a3091d0fc683ff9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c1e1998f44f3232c800201499dab806855e1dce75ff9305ef4096fedbc796014", "generation_seconds_batch": 23.86017801403068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.346304416656494, "target_z": 0.7148436970814094, "ce": 5.5817952156066895, "genuineness": 3.7542223930358887, "identity": 0.783820390701294, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d6ef590efe590f32f6d58c22a4d0f47a47011de7e5ded8d65a3091d0fc683ff9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "d73f0fcb84257f7491aa41db9d4f5ed18ddf43f708334f3eee4a2c73971fbdf3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ee5186d953eefa777edcbfe0f2a689b1fd5c57ccf5412ceeef46e76cb79bba40", "generation_seconds_batch": 21.920783741981722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3313519954681396, "target_z": 0.7031249372276028, "ce": 6.197352409362793, "genuineness": 3.873466968536377, "identity": 0.383059024810791, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d73f0fcb84257f7491aa41db9d4f5ed18ddf43f708334f3eee4a2c73971fbdf3.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": "d76e6f481df69cb8b469bf088cd5cacdebdbb2c22b550ca66710a9326da90db2", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a4c1f15ec1a58479e3dc80d5cc5d211f60bec149f75978254ec005a8b5d39b2", "generation_seconds_batch": 21.12603548099287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0390994548797607, "target_z": 1.2578125299263754, "ce": 6.062326908111572, "genuineness": 3.8546390533447266, "identity": 0.3372887670993805, "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_RANT_reg/d76e6f481df69cb8b469bf088cd5cacdebdbb2c22b550ca66710a9326da90db2.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "d85c2b649a918068cabbe184e36508cfc79e5d071acffa21139dbb5896b5f007", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "531ea02892b774b2a9dacc3af5f02784b8351e37ab0fe6ab3f1e966d914eb409", "generation_seconds_batch": 21.356821218971163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.021160244941711426, "target_z": -1.1406250248169942, "ce": 6.340320587158203, "genuineness": 4.262581825256348, "identity": 0.5386306047439575, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d85c2b649a918068cabbe184e36508cfc79e5d071acffa21139dbb5896b5f007.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "d88da3bd138de47a1a2246db90cd48f8b6b9e2eb40f0ab69abcbf6d7554315cb", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f99cc61d2191707869cdb90a88437e35326e49e6043e3afb1b0719e8eaf1dadd", "generation_seconds_batch": 27.142972182948142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.525733470916748, "target_z": 0.8554688153270877, "ce": 6.038498878479004, "genuineness": 3.6789097785949707, "identity": 0.7479313611984253, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d88da3bd138de47a1a2246db90cd48f8b6b9e2eb40f0ab69abcbf6d7554315cb.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "d896ecd933a4b4e1f39530d49112b345167bcbf7325993927003651f14349986", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "058818968ad4a29f5e03acb5636eddec2c18bdae525f5ed5135b61757be3bacd", "generation_seconds_batch": 23.13543946109712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.044764280319214, "target_z": 0.4785156536490301, "ce": 6.1020402908325195, "genuineness": 3.5094566345214844, "identity": 0.730398416519165, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d896ecd933a4b4e1f39530d49112b345167bcbf7325993927003651f14349986.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "d8db9cd78e630d6e60e39ea98bcf3b0f72f0f03329838a750d13230b94f449dd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ac714f1793cade0165baf0b258c23412c694b55aeb341c1ce95eb32dcdfd9d05", "generation_seconds_batch": 21.77322254993487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8062064051628113, "target_z": -0.49218749343079565, "ce": 5.593709468841553, "genuineness": 3.9927117824554443, "identity": 0.7159380912780762, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d8db9cd78e630d6e60e39ea98bcf3b0f72f0f03329838a750d13230b94f449dd.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "d929ba4f3628249975db5b3039d55e4952ba76a710c58b2190e1b0e6b51fb25f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed77f60342382bf6f3488a5f6d27f21fcb7d234702c1daf135f37c8db3309e9c", "generation_seconds_batch": 22.309256123029627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.889575481414795, "target_z": 1.1406251182456784, "ce": 5.371314525604248, "genuineness": 3.773050308227539, "identity": 0.5636655688285828, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d929ba4f3628249975db5b3039d55e4952ba76a710c58b2190e1b0e6b51fb25f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "d93baf071af27c7f3402eff1e13e175d3f9c6eeec22e4dabc77493da84af3f5f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc2b5766c0335cea6a36aea86e8b92a9e34ccfa6c77ed1f785ae2b90aa9a4ea0", "generation_seconds_batch": 16.84467222506646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0291311740875244, "target_z": 1.250000023357171, "ce": 5.6334228515625, "genuineness": 3.798154592514038, "identity": 0.5416660904884338, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d93baf071af27c7f3402eff1e13e175d3f9c6eeec22e4dabc77493da84af3f5f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "d970d884b9d375837b58087ddb8514f90842a47eba322041700fd49decb89c7a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "714fda3804d442b52507864b5d0ee6193d3da32a9da0ce054e0e838a202ab475", "generation_seconds_batch": 17.646514503052458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1999531984329224, "target_z": -0.18359371751893402, "ce": 5.673136234283447, "genuineness": 4.099404335021973, "identity": 0.6168097853660583, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d970d884b9d375837b58087ddb8514f90842a47eba322041700fd49decb89c7a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "d98340ffe3f68ed19c6da9de39fa711b43e525eaf60045576422a010c506ae46", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f3b99597957c858dec91a94467eaf07507bc4481ec8518852ee4fd4d4bbf29bd", "generation_seconds_batch": 21.57885894703213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.346304416656494, "target_z": 0.7148436970814094, "ce": 6.006728172302246, "genuineness": 3.562803030014038, "identity": 0.6783454418182373, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d98340ffe3f68ed19c6da9de39fa711b43e525eaf60045576422a010c506ae46.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": "d99270529e1caef0928bb5c72ac0cc20f79d396c047c9e295b28d6aeee895cf1", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e56c87324a72595fa90b1001325e2dfbb40abe789a203321fa937127543da338", "generation_seconds_batch": 23.236986513016745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.02595853805542, "target_z": 2.0312501197055015, "ce": 5.557967185974121, "genuineness": 3.5439748764038086, "identity": 0.48362797498703003, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/d99270529e1caef0928bb5c72ac0cc20f79d396c047c9e295b28d6aeee895cf1.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "da0205c44b65366474c84402b6b5211b9fc494fea0c8be925fa9e9ffa95b2859", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "08fec2dce62717ec38d8e6d8841a5504fb63a5b61ddc4aa76bcf39f70500c680", "generation_seconds_batch": 22.854182895971462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6054797172546387, "target_z": 0.9179688678807226, "ce": 5.756534099578857, "genuineness": 3.515732765197754, "identity": 0.141611248254776, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/da0205c44b65366474c84402b6b5211b9fc494fea0c8be925fa9e9ffa95b2859.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "da0966ea94d658af2d2368541ffb269bd47c221d11232013ab7a7171f836a6c7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f8d0319ac5a90649bd83139f410987e64962b78015886d21687ee154b6bdc5d0", "generation_seconds_batch": 26.17832922504749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8397340774536133, "target_z": 1.1015625853996567, "ce": 5.998785495758057, "genuineness": 3.8044304847717285, "identity": 0.36602357029914856, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/da0966ea94d658af2d2368541ffb269bd47c221d11232013ab7a7171f836a6c7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "da4a9f68722fcaeba45c9c2284fe373e82321cd170cb83f86812503f33753944", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "56e071e7f700af7dff69f462c89baf1afbeb65807b13eb5408d9bb380858b8c5", "generation_seconds_batch": 18.55227396299597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07852256298065186, "target_z": -1.0624999591249507, "ce": 5.919358730316162, "genuineness": 4.118232727050781, "identity": 0.783735454082489, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/da4a9f68722fcaeba45c9c2284fe373e82321cd170cb83f86812503f33753944.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "da68bb5b434d2bf4f849494eb36759f44e3e816a3394d92e133c0434bd57b380", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9455287600f741582c5b7ad800f2a8f78042dc5967f919ff4602ed50c3754c27", "generation_seconds_batch": 17.301613447023556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5175766944885254, "target_z": 1.632812658390816, "ce": 5.645336627960205, "genuineness": 3.7542223930358887, "identity": 0.5249835848808289, "hypothesis": "J'ai trouvé la 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_RANT_reg/da68bb5b434d2bf4f849494eb36759f44e3e816a3394d92e133c0434bd57b380.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "daf3e714da0ec30f9b2a305df17881cf3da623bbd5237f567019f9f86992463f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4e23294457ff01a6d98f3ef54b970c9a7b7031386a341b55691ddb5d3c20c042", "generation_seconds_batch": 19.130376418004744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.929448366165161, "target_z": 1.1718749576651275, "ce": 5.839932441711426, "genuineness": 3.9173994064331055, "identity": 0.2517988681793213, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/daf3e714da0ec30f9b2a305df17881cf3da623bbd5237f567019f9f86992463f.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "e2c2b0470da62de76a655c6c5eaf2c8e9a4a5cde1571ed52abcf8bef905419ae", "generation_seconds_batch": 15.861990939010866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6367457509040833, "target_z": -0.6250000116785855, "ce": 6.419747352600098, "genuineness": 4.18099308013916, "identity": 0.40057289600372314, "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_RANT_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "db29b510abb7a9b66e42ae21f9ec5745238b09e00d1621769531dce79ec70b98", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de3c7542557fbb34de20207b7410ddc93899cdcc4dfb0105c8c140a9f158fb08", "generation_seconds_batch": 22.287493029027246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4858603477478027, "target_z": 0.8242187890502704, "ce": 5.80021858215332, "genuineness": 3.6663577556610107, "identity": 0.5040315389633179, "hypothesis": "I found the little box! It was beside the old window! We can look inside together!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/db29b510abb7a9b66e42ae21f9ec5745238b09e00d1621769531dce79ec70b98.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": "db9c0a796ad7c8a2a267343026cc3719cc0723c0b99e7dae70f234f05d126f98", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a2773728d1b0c023489f2ed2826d5767f037a45e6d168bc1a978c086ff38a78", "generation_seconds_batch": 20.74382221198175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.799860954284668, "target_z": 1.0703125591228393, "ce": 6.0146708488464355, "genuineness": 3.7542223930358887, "identity": 0.7029761075973511, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de laag. En daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/db9c0a796ad7c8a2a267343026cc3719cc0723c0b99e7dae70f234f05d126f98.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": "dc3e8d996fdd567d4e943025f85135c8aee5aea8e79874fa382303abb4f7a504", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9f418f17cb21c729b664bf1b04d9ff2c2c1e8a7aa6d84d696991683934804d5", "generation_seconds_batch": 21.256950397975743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.258401393890381, "target_z": 1.4296874875915029, "ce": 6.173524379730225, "genuineness": 3.9048471450805664, "identity": 0.393593430519104, "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_RANT_reg/dc3e8d996fdd567d4e943025f85135c8aee5aea8e79874fa382303abb4f7a504.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "ddf6319dd972ef5ce466172c966a774ff4a8caf0b0dea5f3b11a30e7f9186ac9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "5979a267ae045b994873747b58b544a366421cf4abac104594b985bbb672cc30", "generation_seconds_batch": 28.15482666902244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.281959533691406, "genuineness": 3.653805732727051, "identity": 0.6735010147094727, "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_RANT_reg/ddf6319dd972ef5ce466172c966a774ff4a8caf0b0dea5f3b11a30e7f9186ac9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "de1e96d5efe4d59061265c8892bfaef4fc4725df15498c58ea4bd315b608266f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d96168490e9e832cd4c0ad2f8c0aa035e6cc03b0e1981c717961a70a65a970ff", "generation_seconds_batch": 20.831367084058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6702733039855957, "target_z": 0.9687499737231826, "ce": 5.792276382446289, "genuineness": 3.528284788131714, "identity": 0.56684809923172, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/de1e96d5efe4d59061265c8892bfaef4fc4725df15498c58ea4bd315b608266f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "de3591ec0d8b0d53d30169c71314ab8a0f6dbb7a403ae511ee70cd456de9057a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "156da83c7ae5c875240a6fa5220cef53c0c5b598ef4e445a697102b6dd2afaff", "generation_seconds_batch": 16.793130935984664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9262757301330566, "target_z": 1.953125054013458, "ce": 5.33954381942749, "genuineness": 3.6192874908447266, "identity": 0.5698862671852112, "hypothesis": "¡La puerta sigue abierta! ¡He dejado tu abrigo sobre la silla! ¡Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/de3591ec0d8b0d53d30169c71314ab8a0f6dbb7a403ae511ee70cd456de9057a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "de71194511171ca2e129268e24c3ef2c793c1818a760b4bad22caed85581af51", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d43e67d293be233dd3a17d93e6c516992aa3ba1f2b46b4943208a7544b7f59aa", "generation_seconds_batch": 24.230891890008934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1587185859680176, "target_z": 1.3515624218994593, "ce": 5.7128496170043945, "genuineness": 3.5314228534698486, "identity": 0.5484251379966736, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/de71194511171ca2e129268e24c3ef2c793c1818a760b4bad22caed85581af51.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": "ded0222f0df9df74ad5d4bcafaea076aaecc49e87b591a49c3cfcaabcb444a47", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "966b6bb8c24c1abc0cadf406464be42d59e8334cb9fbae8042fef0338a37edb5", "generation_seconds_batch": 26.506481830962002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2316694259643555, "target_z": 0.6250000583929276, "ce": 5.990842819213867, "genuineness": 3.47493839263916, "identity": 0.5736569166183472, "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_RANT_reg/ded0222f0df9df74ad5d4bcafaea076aaecc49e87b591a49c3cfcaabcb444a47.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "df25fbbbec165ba7088e9956773c8b4b41c10f5f7f4402b68bf39af21ad79209", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff433f7015101af7d3db7fe852bbd8f744b1c74a6bcf80e41067dd8ddcd08cc5", "generation_seconds_batch": 22.49244651698973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.72691011428833, "target_z": 1.796874922629371, "ce": 5.49045467376709, "genuineness": 3.6789097785949707, "identity": 0.5136232972145081, "hypothesis": "La puerta sigue abierta! He dejado tu abrigo sobre la silla! Dime cuándo estás listo!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/df25fbbbec165ba7088e9956773c8b4b41c10f5f7f4402b68bf39af21ad79209.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "df60ca9d6ea7e463f20f80788c49503771fb67020b1a1f231717aa6d481a96af", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e31a7fbedbcb96efc7d50be5d0924bf7cba7dec888ee5ff92fe104df48964dfb", "generation_seconds_batch": 22.966856123995967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36760246753692627, "target_z": -0.8359374554753927, "ce": 6.284721851348877, "genuineness": 4.118232727050781, "identity": 0.5571776628494263, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/df60ca9d6ea7e463f20f80788c49503771fb67020b1a1f231717aa6d481a96af.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": "df6ea218460715583660d9a17d6e2475b4bfa57b3249ea907593db1d998da41c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "880448ed600921667341dccf063983e3d1f264bbf027db4ce00c7f2f12d36646", "generation_seconds_batch": 16.31859594595153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8864026069641113, "target_z": 1.9218750277366405, "ce": 5.982900142669678, "genuineness": 3.465524435043335, "identity": 0.47742557525634766, "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_RANT_reg/df6ea218460715583660d9a17d6e2475b4bfa57b3249ea907593db1d998da41c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "df7f3daff4479aca4739d2579c871695b8c4b3a5a915410864a40be6e0d5cc1e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e9a9be4b89725c9d2a864a7327637469a2cb08a0821aabd40545b990bce2d8da", "generation_seconds_batch": 24.072464816039428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.387989044189453, "target_z": 1.5312500729911596, "ce": 5.557967185974121, "genuineness": 3.3462796211242676, "identity": 0.45779502391815186, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/df7f3daff4479aca4739d2579c871695b8c4b3a5a915410864a40be6e0d5cc1e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "dff1e218d5987ffe9894c015d8b624d67d05587f7fa0a0e227af26d0338d24d5", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "59dab541ce1294e905b82aaee3d07d7c07b9437af8f7a65d94905c6e26b378c5", "generation_seconds_batch": 18.07022723800037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2423182725906372, "target_z": -0.15039065802849969, "ce": 6.173524379730225, "genuineness": 3.71028995513916, "identity": 0.27622246742248535, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/dff1e218d5987ffe9894c015d8b624d67d05587f7fa0a0e227af26d0338d24d5.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": "e0137694250c4b6b37e39eac0ed95333e8998c9bbe50a7f1e39300e47e1c0d1c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31ccf07728e53777e7278cf71b5eda1a425e9d68a65ee59e44e0878cedd5daa3", "generation_seconds_batch": 16.260466612991877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.02595853805542, "target_z": 2.0312501197055015, "ce": 5.744620323181152, "genuineness": 3.6914620399475098, "identity": 0.3544779419898987, "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_RANT_reg/e0137694250c4b6b37e39eac0ed95333e8998c9bbe50a7f1e39300e47e1c0d1c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "e077d5fd1857eee8cc88645fae00cbdc8ffb8273b8baa4028d2cb57bce1c13c5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79e4646a71abbb4c6ac6ad715fdfbf3920be557ee2febdc668c419f64be12df3", "generation_seconds_batch": 25.471652873093262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0590360164642334, "target_z": 1.2734375430647842, "ce": 5.629451274871826, "genuineness": 3.6412534713745117, "identity": 0.6426194310188293, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e077d5fd1857eee8cc88645fae00cbdc8ffb8273b8baa4028d2cb57bce1c13c5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "e0889924b03af9f83966e20a6d5a648c6cd4653e423c16559d4c2c2e44223335", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f9c16c25a10ad709149af84a986f82fd1085e2e35f65625e08b43a821c0597f", "generation_seconds_batch": 22.64774499705527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8229701519012451, "target_z": 0.30468747591291734, "ce": 6.054384231567383, "genuineness": 4.030367851257324, "identity": 0.8021875023841858, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e0889924b03af9f83966e20a6d5a648c6cd4653e423c16559d4c2c2e44223335.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "e0b85c61e9e9d550b0d4a02a36b4a941c976487d6917180df99d280368de6caf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6d2fe3ff1ded554aa27d6128b5ac35b0f71d86001208b371d96a6e1710272894", "generation_seconds_batch": 18.874024434015155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4370111227035522, "target_z": 0.002197255794003205, "ce": 5.998785495758057, "genuineness": 3.8044304847717285, "identity": 0.6507565975189209, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e0b85c61e9e9d550b0d4a02a36b4a941c976487d6917180df99d280368de6caf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "e15c94a3019c2995d0712a47fd71a3d7e5c9190aedcd0ff37436a3a197dbb178", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "724bdf598d01a1b035a7c2903bd14e5c6a3a588d0d05cf8e03a252acde68efce", "generation_seconds_batch": 16.064549545990303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1188454627990723, "target_z": 1.320312395622642, "ce": 5.816103935241699, "genuineness": 3.578493118286133, "identity": 0.5570864677429199, "hypothesis": "I found the little box! It was beside the old window! We can look inside together!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e15c94a3019c2995d0712a47fd71a3d7e5c9190aedcd0ff37436a3a197dbb178.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "e1b30f7d5aba20329faa3331a247671462c26758073d15983b00612669cc7a43", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d02a9e98407bcb21cbcbe04c72f6857fc78874cf0dbe0263df6b6dbffc9ddf0a", "generation_seconds_batch": 23.563897556974553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.241637706756592, "target_z": 0.632812564962132, "ce": 6.229123115539551, "genuineness": 4.149612903594971, "identity": 0.5068399906158447, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e1b30f7d5aba20329faa3331a247671462c26758073d15983b00612669cc7a43.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "e203e667bc52c208af40132c0ae6f99110ffc083252dc558e20e5320fc551500", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "352a1e851cf9e5fcb7b564f8f525c2b0efaa45d223c9f6ce36876c60a51d0a83", "generation_seconds_batch": 24.72982134995982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3512885570526123, "target_z": 0.7187499503660115, "ce": 6.086154937744141, "genuineness": 3.6789097785949707, "identity": 0.7572503685951233, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e203e667bc52c208af40132c0ae6f99110ffc083252dc558e20e5320fc551500.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "e205920078f9f9c01596c5ecb32f933318c1d36f90d5cfcba4bf338bed9f3c12", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "71fbeec5d35258efe1a64d05411d398d20f0e621b76d3f64670b2fdbaa73eca2", "generation_seconds_batch": 19.885150048998185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6168092489242554, "target_z": -0.6406249781026522, "ce": 6.41180419921875, "genuineness": 4.061748504638672, "identity": 0.38075941801071167, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e205920078f9f9c01596c5ecb32f933318c1d36f90d5cfcba4bf338bed9f3c12.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "e20d5a13dd16f39609549b1c7b1e7f8e91453bafa684930d0e6ab00f15550976", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ae3c90b41d11788e2e54e809ea8cc4e14c583e610b3737b70a6c72423d9bb2a8", "generation_seconds_batch": 25.881722176913172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.5659098625183105, "genuineness": 3.4435582160949707, "identity": 0.39305242896080017, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e20d5a13dd16f39609549b1c7b1e7f8e91453bafa684930d0e6ab00f15550976.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "e28ffea27ada909d41e6f9471bdd1ebee176e95fe189f3cebae9d49e5972e7c4", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "80b41998ba0f7c8647cc79f3f8096d7ab8ae444e288270d6d05d38d65499ed1a", "generation_seconds_batch": 15.841209507081658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.16551399230957, "target_z": 2.140624837959626, "ce": 5.80021858215332, "genuineness": 3.20820689201355, "identity": 0.32647329568862915, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e28ffea27ada909d41e6f9471bdd1ebee176e95fe189f3cebae9d49e5972e7c4.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": "e3396ef8d620d8982fd5553af724b5f4dfe1f535b70e45bcff5ab0ab1e89222e", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae4509d1b3bf08af4920b0c846842c2962869d49e0fe0c88eff78a916c6d20f2", "generation_seconds_batch": 20.688292101956904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5275449752807617, "target_z": 1.6406251649600205, "ce": 5.792276382446289, "genuineness": 3.5816311836242676, "identity": 0.6989887952804565, "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_RANT_reg/e3396ef8d620d8982fd5553af724b5f4dfe1f535b70e45bcff5ab0ab1e89222e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "e3506f9f45bf1daa44a786f98dc67d03cdb5f1c8b66f710278b56b5eac702900", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d876d6e4333ae97e66c5be8a25a66667983bf645ad19c2fa35f2b248ee0316ee", "generation_seconds_batch": 22.827915480011143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.078972339630127, "target_z": 1.2890623693458245, "ce": 5.752562999725342, "genuineness": 3.575355052947998, "identity": 0.2589743137359619, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e3506f9f45bf1daa44a786f98dc67d03cdb5f1c8b66f710278b56b5eac702900.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "e39f085010bfee8b4dee3b172c6f0d234bc1b0d5a6be31ec6d989bab121eecb6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bf2fab49e359b94700aad790250b160383fe5c96446620677ab9aa33265cd90a", "generation_seconds_batch": 23.702725843992084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9226529598236084, "target_z": 0.3828125416049609, "ce": 5.959072113037109, "genuineness": 3.5690791606903076, "identity": 0.7210655808448792, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e39f085010bfee8b4dee3b172c6f0d234bc1b0d5a6be31ec6d989bab121eecb6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "e3a1b81dbe0a27c961bb174fb16ca7f2ffbc26778e5e63a867d0f0b1b383b803", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4b81e426c21b35ba8fc16cd7e9bc674eb38b66edf43b670b66f4e03e9d81c56e", "generation_seconds_batch": 23.11292018101085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6161285638809204, "target_z": 0.1425781514592953, "ce": 6.213237762451172, "genuineness": 3.9425034523010254, "identity": 0.6017034649848938, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e3a1b81dbe0a27c961bb174fb16ca7f2ffbc26778e5e63a867d0f0b1b383b803.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "e3f794247618846896bf6ed82e83fe72ab454ec29c205e43136091fdfdc1a982", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "c4ce604a0d3f2793fdfb4425276ccb45209c64cbe28faec0a0dee9cf03c58181", "generation_seconds_batch": 19.756685576052405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018712878227233887, "target_z": -1.1093749985401768, "ce": 6.229123115539551, "genuineness": 4.381826400756836, "identity": 0.35630106925964355, "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_RANT_reg/e3f794247618846896bf6ed82e83fe72ab454ec29c205e43136091fdfdc1a982.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": "e439116c21cd1804e2df3b3a60bf1040417d4dd6fe85cb9b72af814246b9957a", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0eeb443ea124b7fd6c9c9fb0e46dae4d79783a872ad45f7d4f18e4c1763e3f16", "generation_seconds_batch": 22.73466582398396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.560622215270996, "target_z": 0.8828124014619346, "ce": 6.1020402908325195, "genuineness": 3.562803030014038, "identity": 0.6496763229370117, "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_RANT_reg/e439116c21cd1804e2df3b3a60bf1040417d4dd6fe85cb9b72af814246b9957a.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": "e45942e2d3e3a7468f88edfaa8b3126460ded96aca068109c27cebade0cb7f3b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b9d328ae5f9901341a01b8a6f18cd1fd45638bb82aba8db1e216634a508132c2", "generation_seconds_batch": 20.736126290052198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7300829887390137, "target_z": 1.0156250131384088, "ce": 5.863759994506836, "genuineness": 3.8420867919921875, "identity": 0.5981031656265259, "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_RANT_reg/e45942e2d3e3a7468f88edfaa8b3126460ded96aca068109c27cebade0cb7f3b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "e4ad6b6ecdec83cb8f33e9a1746b4129ea71b2c1edf4937158f10e9ccc8a11e6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "75b98ac09a4dc08852cb5e63fbf98083daca6a5b7027cafeee59b2f32fe9f64f", "generation_seconds_batch": 22.838239355012774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1376514434814453, "target_z": -0.23242188357646124, "ce": 5.871703147888184, "genuineness": 4.218649387359619, "identity": 0.5975421667098999, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e4ad6b6ecdec83cb8f33e9a1746b4129ea71b2c1edf4937158f10e9ccc8a11e6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "e4b2cc2404a187443963e0e5e0ef986240dc6986e38e3577a885083e785eb89b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c9c39aca93c81eeff7283521ef6849f9af12f91fab59125cd24d4c67b946d996", "generation_seconds_batch": 22.083597937016748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34268176555633545, "target_z": -0.8554687218984036, "ce": 6.252951145172119, "genuineness": 4.507347106933594, "identity": 0.7998682260513306, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e4b2cc2404a187443963e0e5e0ef986240dc6986e38e3577a885083e785eb89b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "e538f1296c74fcd3a31449db306d7c35183f404099eeaa7cb56a3fa3e904aaf5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4732f412c5365f77fa1b9818b614461e4dc944662e9b4a045be9ed18f9811b99", "generation_seconds_batch": 21.805613231961615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0304925441741943, "target_z": -0.3164062357667239, "ce": 6.268836498260498, "genuineness": 4.545003414154053, "identity": 0.5836392641067505, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e538f1296c74fcd3a31449db306d7c35183f404099eeaa7cb56a3fa3e904aaf5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "e5f69bc6b3bf527dcf4f35c4dd2d08386b034a0525d52aca0f166ae1cc3f54f6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "248a7feefe62403136c6f475c1412884cd11fc0f3a1c6bbf23ab22be21cc3434", "generation_seconds_batch": 21.567104470916092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.46456241607666, "target_z": 2.3750000350357565, "ce": 5.692992687225342, "genuineness": 3.5722172260284424, "identity": 0.22939661145210266, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e5f69bc6b3bf527dcf4f35c4dd2d08386b034a0525d52aca0f166ae1cc3f54f6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "e683c333d328524c948705b6a6b6d9d2b17190d3876678af3eee71840045f39e", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b41d8d6143b5237fdae253f08f58f855427311056168ee878d9b2bf9ceb5352a", "generation_seconds_batch": 19.76697251398582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1737864017486572, "target_z": -0.20410159397743755, "ce": 6.260893821716309, "genuineness": 3.9550557136535645, "identity": 0.2630332112312317, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e683c333d328524c948705b6a6b6d9d2b17190d3876678af3eee71840045f39e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "e6c0db587f83cc9f09eeca14c285ee02ec1b39d9f379ba5244e8a8d4997b8035", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "cc1bee7aa7d4cb51b41fb2f4b206ae451e6cf020e6fa42b9473fe4e2ff292f38", "generation_seconds_batch": 21.57885894703213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2914791107177734, "target_z": 0.6718750978081537, "ce": 6.125868320465088, "genuineness": 3.8169827461242676, "identity": 0.553803026676178, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e6c0db587f83cc9f09eeca14c285ee02ec1b39d9f379ba5244e8a8d4997b8035.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "749b67d25ab6e7b74cc311787147407daa2b81467ba740d6ba1ec611d942f6f0", "generation_seconds_batch": 15.66625829495024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4872216582298279, "target_z": -0.7421875167879667, "ce": 6.300607204437256, "genuineness": 4.281409740447998, "identity": 0.47741708159446716, "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_RANT_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.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": "e7263a4bf981345998c5bdf51a3c62abbac60ea660be63c625bced994405116a", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d61aab4ab4d1e1795265e0076eddd4d2b89d2f1742cccbade33c6b2b1291af3d", "generation_seconds_batch": 16.955784159014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1288137435913086, "target_z": 1.3281249021918462, "ce": 5.784333229064941, "genuineness": 3.798154592514038, "identity": 0.7341997027397156, "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_RANT_reg/e7263a4bf981345998c5bdf51a3c62abbac60ea660be63c625bced994405116a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "e783e74aba65f452f47758d756c3060697660c59f65da0f8f5749c367880393a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "30464a883cf22138c7009b5d4e6c155b60769cd5bdffbbe4a5fff840410fb6d5", "generation_seconds_batch": 22.21559696807526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7830970287322998, "target_z": 0.27343744963609995, "ce": 6.109982967376709, "genuineness": 4.218649387359619, "identity": 0.6903079152107239, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e783e74aba65f452f47758d756c3060697660c59f65da0f8f5749c367880393a.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": "e7b948c218f45ca276075836d992724ab33af359f4c79499fda675ce842c5a44", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8268db43b54bb1a50bcbdb27f79c7e2b6ec382eb8d358f76f05c2017c76bbf01", "generation_seconds_batch": 15.8070251760073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.225323677062988, "target_z": 2.187499877374852, "ce": 5.8875885009765625, "genuineness": 3.528284788131714, "identity": 0.5891544818878174, "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_RANT_reg/e7b948c218f45ca276075836d992724ab33af359f4c79499fda675ce842c5a44.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "e7f236abcbec3ab8d9bbb114ab7beff0282fa63af16d6b3a088e4def2c78830a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "7b9e0697593279441cda685161bcdc42587dc9261a37a72480995120ee2b6b24", "generation_seconds_batch": 22.98833900608588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9525578022003174, "target_z": 0.40625006131257396, "ce": 6.062326908111572, "genuineness": 3.9927117824554443, "identity": 0.6792964339256287, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e7f236abcbec3ab8d9bbb114ab7beff0282fa63af16d6b3a088e4def2c78830a.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "e82f7cc853c5a517bb6f1e5a89e320b816ca332bae95c00480183a92ace7cdb7", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "623a1060f9cbf846f675bd54b9705339d171c2e96a894cc3568224ad22d9d6fd", "generation_seconds_batch": 16.769868962932378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8696389198303223, "target_z": 1.1250001051072698, "ce": 5.577824115753174, "genuineness": 3.7667744159698486, "identity": 0.6560122966766357, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¡Podemos mirar juntos lo que hay dentro!", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e82f7cc853c5a517bb6f1e5a89e320b816ca332bae95c00480183a92ace7cdb7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "e859135a0a60adef5cd267a394809d91587cb728ab83df7a0cd2ff1954840772", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ed5eed9ca3b3d6e8c6ebf74e5763c446f04ead6109eb690569183ba96a51d10a", "generation_seconds_batch": 22.949870598036796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5805587768554688, "target_z": 0.8984374146003434, "ce": 6.09409761428833, "genuineness": 4.005263805389404, "identity": 0.6569862365722656, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/e859135a0a60adef5cd267a394809d91587cb728ab83df7a0cd2ff1954840772.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "c3ae604877e69e13c37201009c6852ca4e6d43438830092c2912a076e77471b8", "generation_seconds_batch": 15.248356981086545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.13081121444702148, "target_z": -1.226562503649558, "ce": 6.427689552307129, "genuineness": 4.243753433227539, "identity": 0.5633001923561096, "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_RANT_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.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": "e9f6b99e5ca8a697f3ca2a2963a00202ffc572a0583177adeddca731ab6f68ca", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "446ae8b0b3619cf75f5f4b07e8fc18f1ee196782b5bc881a8dd8b2bf7d222904", "generation_seconds_batch": 20.736126290052198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7799243927001953, "target_z": 1.0546875459844305, "ce": 6.046441555023193, "genuineness": 3.898571252822876, "identity": 0.595514714717865, "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_RANT_reg/e9f6b99e5ca8a697f3ca2a2963a00202ffc572a0583177adeddca731ab6f68ca.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "ea08502e643f770ae8cd030d50ab3b155dce167c92bc1842ff3c4f3055311592", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "38a1b814830dd0a452cd5b90b5bb8dd7284f63172912a8f598fafc5e4e7befca", "generation_seconds_batch": 24.958994923043065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627227306365967, "target_z": 1.7187498569373274, "ce": 5.418970584869385, "genuineness": 3.7416701316833496, "identity": 0.27190664410591125, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ea08502e643f770ae8cd030d50ab3b155dce167c92bc1842ff3c4f3055311592.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "ea947507537b4d4eecb1cf45f1e8aab4196b2c9d2031dcceecccfd4c6e5e2f75", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "76fcb9917cacbe2d7415adf1f27eae2d6dd057d79e57c44b059997aaae00a29b", "generation_seconds_batch": 17.501822380931117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8696389198303223, "target_z": 1.1250001051072698, "ce": 6.1894097328186035, "genuineness": 3.8671910762786865, "identity": 0.5925505757331848, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ea947507537b4d4eecb1cf45f1e8aab4196b2c9d2031dcceecccfd4c6e5e2f75.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "eb13092acb1917ebc158325a3ff4e0b1a8795a82c3d4e0f34a4b4a132ce548ec", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a8c8d5898051037bce130d9a45f80ca561213b6357c56dd764b08796556d9be0", "generation_seconds_batch": 20.616082031978294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4523327350616455, "target_z": -0.7695312430658399, "ce": 6.38003396987915, "genuineness": 4.168440818786621, "identity": 0.5975536704063416, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/eb13092acb1917ebc158325a3ff4e0b1a8795a82c3d4e0f34a4b4a132ce548ec.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "eb76db2399a0e6133a3922e7c10fe0a4f3f360e0267104bb2e34498a60688f3a", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "fa2ad73cfc10acdfd6850f6c16ab722a9efb88bd16051ea39ee4c99252a50ec0", "generation_seconds_batch": 16.971062553930096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17820525169372559, "target_z": -0.9843749868615913, "ce": 6.0146708488464355, "genuineness": 4.645420074462891, "identity": 0.7405415773391724, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/eb76db2399a0e6133a3922e7c10fe0a4f3f360e0267104bb2e34498a60688f3a.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "c605478554da67afd3a96a09605ccdeed5fcdce2f2207ff871ba255c7c11817c", "generation_seconds_batch": 14.691593277966604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2778879404067993, "target_z": -0.9062500145982318, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.7254563570022583, "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_RANT_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.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": "ebf6a0b66f0e85e4edf04a68030cb747d7b300e3c0b03f19092af220c0e628d6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc86b4567a682c29a9e89f1982e3f2cd84457bff1141b4f06379d9e9ab9d5438", "generation_seconds_batch": 24.671221949975006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5955114364624023, "target_z": 0.9101563613115182, "ce": 5.585766792297363, "genuineness": 3.8922953605651855, "identity": 0.4395475387573242, "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_RANT_reg/ebf6a0b66f0e85e4edf04a68030cb747d7b300e3c0b03f19092af220c0e628d6.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": "ec409015d0105081219523a02eefb7911bb589488f1b319abee39c46a109786b", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4968c4241b5ecfd2f25ea8109fa2423b5da0a093ab686d9b33b4413fe90fb617", "generation_seconds_batch": 25.42512089491356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.268369674682617, "target_z": 1.4374999941607072, "ce": 6.109982967376709, "genuineness": 3.7165660858154297, "identity": 0.37903541326522827, "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_RANT_reg/ec409015d0105081219523a02eefb7911bb589488f1b319abee39c46a109786b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "ec80751d59bac4f1aed0e629f610a830263cf2893b58be528bba48407f008f4b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "35923d9bc36312321ac0a40d3f52f74183f586ee3fe1e76816f1155d2ae819a2", "generation_seconds_batch": 26.067431767005473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.74005126953125, "target_z": 1.023437519707613, "ce": 5.692992687225342, "genuineness": 3.4623863697052, "identity": 0.4228079617023468, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ec80751d59bac4f1aed0e629f610a830263cf2893b58be528bba48407f008f4b.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": "ecd066735a02816eab6b9088ee03b8b84b0c4efe5faceb6f82d92237fba03917", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eeee9b06bbb2cb30a77c325b96b6b120542fb2c7cd016a92c05f34c5196ba4f3", "generation_seconds_batch": 16.78427623200696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.929448366165161, "target_z": 1.1718749576651275, "ce": 5.80021858215332, "genuineness": 3.503180742263794, "identity": 0.2873138189315796, "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_RANT_reg/ecd066735a02816eab6b9088ee03b8b84b0c4efe5faceb6f82d92237fba03917.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "ed2bbc4aa07d90b61e0ddb4cba7298c571229044e9832e6518b113176c40845f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "9534d69b7a30c0079091071f28a8f564b922e1fc95cc96ab1525d32127b037be", "generation_seconds_batch": 21.805613231961615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.217397689819336, "target_z": -0.1699218310228264, "ce": 6.229123115539551, "genuineness": 4.111956596374512, "identity": 0.4116896390914917, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ed2bbc4aa07d90b61e0ddb4cba7298c571229044e9832e6518b113176c40845f.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": "ed4271cc4f716151c720321dc1393cfb99cb90aa55e6babf16c546fb48fd7164", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a69e012f77ab4a5d79b5c5f2d7defc50d02044a13d23d6ae378a1360a07734c", "generation_seconds_batch": 20.688292101956904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3313519954681396, "target_z": 0.7031249372276028, "ce": 5.736677646636963, "genuineness": 3.8169827461242676, "identity": 0.6318615674972534, "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_RANT_reg/ed4271cc4f716151c720321dc1393cfb99cb90aa55e6babf16c546fb48fd7164.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": "ed8759a39177a348903843ba581a22f63f50f9933dcd477823656b4ffd74dd49", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a7a36717774679d20167bd9a65be2719302d56fe2ac51d22e17ef1bec9d972d", "generation_seconds_batch": 21.28093650494702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348115921020508, "target_z": 1.5000000467143422, "ce": 5.998785495758057, "genuineness": 3.6851859092712402, "identity": 0.5597201585769653, "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_RANT_reg/ed8759a39177a348903843ba581a22f63f50f9933dcd477823656b4ffd74dd49.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "ede6c9687d43cc98015d59a1b29a57b77f4c5e29cd95fb16fb6ebf6306d2b8cd", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f095b9e9e7f0e7784a612261c4fb1c818b6c2a3ade16e38bb44d8b95807743e7", "generation_seconds_batch": 18.297362613026053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9992263317108154, "target_z": 1.226562503649558, "ce": 5.77639102935791, "genuineness": 3.5941832065582275, "identity": 0.51956707239151, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ede6c9687d43cc98015d59a1b29a57b77f4c5e29cd95fb16fb6ebf6306d2b8cd.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "ee6b471c09c87ccc730d6de79606836653a7996d4eae09a1352a11cdc03a442a", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "76cb96f15cd2abfa89a465979538c2de9c2658fcf25906f74eaf3f9c08784912", "generation_seconds_batch": 21.314187624957412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1483004093170166, "target_z": -1.0078125065692043, "ce": 6.022613525390625, "genuineness": 4.494794845581055, "identity": 0.43207889795303345, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ee6b471c09c87ccc730d6de79606836653a7996d4eae09a1352a11cdc03a442a.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": "ee7d5a347d2ce3b6d7dd6a4398218d8595ce2c0f0e87f5329f61229627d115c5", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a479b4d6c0362f505d9aebfee85d3456afd0db9d9fdb1f118e082a5c3a3453c6", "generation_seconds_batch": 25.612769053084776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.02595853805542, "target_z": 2.0312501197055015, "ce": 6.157639026641846, "genuineness": 3.7793264389038086, "identity": 0.35340073704719543, "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_RANT_reg/ee7d5a347d2ce3b6d7dd6a4398218d8595ce2c0f0e87f5329f61229627d115c5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "ef6c21abdfa6186b74445a4d09f93ca030509d0f57a9f242bc82d52f1a3ab3a5", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef346326a668f3b7fb6a1554f4bfedfa3259f2febd7f2b065e1c37e217647d4b", "generation_seconds_batch": 26.17832922504749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8397340774536133, "target_z": 1.1015625853996567, "ce": 6.062326908111572, "genuineness": 3.722842216491699, "identity": 0.16026678681373596, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ef6c21abdfa6186b74445a4d09f93ca030509d0f57a9f242bc82d52f1a3ab3a5.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "f03836e2f62a151c0660ea2838ee9f5a089b0402cdb56324f5cc11d35c26b09f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "227a32ac6390b1b7a158f20a720693e67bb7e376b6ece44c29730059104701f0", "generation_seconds_batch": 21.50867951498367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.13081121444702148, "target_z": -1.226562503649558, "ce": 6.23706579208374, "genuineness": 4.582659721374512, "identity": 0.5609495639801025, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f03836e2f62a151c0660ea2838ee9f5a089b0402cdb56324f5cc11d35c26b09f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "f0384382091e9fc63c86e40aaa9cff84388646cf365282739c5439ae70fbb0af", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d35a21c8e0d84d8e59406182aade612a2a0d747a768334ea4bf973b8d2bea473", "generation_seconds_batch": 15.08883035206236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4852949380874634, "target_z": 0.04003904945283024, "ce": 6.308549880981445, "genuineness": 4.1998209953308105, "identity": 0.3838229179382324, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f0384382091e9fc63c86e40aaa9cff84388646cf365282739c5439ae70fbb0af.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "f038e9fa577b9683019b54d4104651b77f0e87259fec14080dfdb11ed2bde14f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e9e517874d1646aa07828c6542e265675bc35bd90567c03e7798b9fb3d469d6d", "generation_seconds_batch": 16.793130935984664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.16551399230957, "target_z": 2.140624837959626, "ce": 5.216432571411133, "genuineness": 3.4310061931610107, "identity": 0.6977800130844116, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f038e9fa577b9683019b54d4104651b77f0e87259fec14080dfdb11ed2bde14f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "f0d43495beabf825c25b70cb74e954019aafcb58979c540b1677a003e11cbe1b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "362169f54bf6e9c54899a80c330ff8df88ef2871b577cf87eb5800d394f93c11", "generation_seconds_batch": 22.234727873001248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31776106357574463, "target_z": -0.8749999883214145, "ce": 6.467403411865234, "genuineness": 4.068024158477783, "identity": 0.551424503326416, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f0d43495beabf825c25b70cb74e954019aafcb58979c540b1677a003e11cbe1b.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": "f11ee34d528d511f9aa214db95a0491e9f7351067d8347bf1cf755d3117892ba", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7b1d0a755a625fc755e161ea76c1bbb3f0f22482c840d796e0c8aca01f96e7e", "generation_seconds_batch": 16.842396187013946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8995437622070312, "target_z": 1.1484376248148827, "ce": 5.780362129211426, "genuineness": 3.6914620399475098, "identity": 0.6519303321838379, "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_RANT_reg/f11ee34d528d511f9aa214db95a0491e9f7351067d8347bf1cf755d3117892ba.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "f18f847c2d3d4ef23e0c23d812c7ff40b3a19d94f2f6d0ef7b7fb352b5d76fce", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c910b3fd3231e953dcfa23c34437d25ab9591fb151243d2c43c1e5e6db5b02d4", "generation_seconds_batch": 22.5495894550113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4559555053710938, "target_z": 0.8007812693426573, "ce": 6.197352409362793, "genuineness": 3.5941832065582275, "identity": 0.5296154618263245, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f18f847c2d3d4ef23e0c23d812c7ff40b3a19d94f2f6d0ef7b7fb352b5d76fce.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-0", "key": "f20bb825c12cc1117ab85fc7b972a4b3adb16f3232734d8b9ed13bd0b7866b90", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3a653216610a484e85f1b5f12fa9ef91cf373186321da01f12fdf3ebb6bdf5f0", "generation_seconds_batch": 25.269430187996477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.019162893295288, "target_z": 1.2421875167879668, "ce": 5.77639102935791, "genuineness": 3.5690791606903076, "identity": 0.647983193397522, "hypothesis": "The 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_RANT_reg/f20bb825c12cc1117ab85fc7b972a4b3adb16f3232734d8b9ed13bd0b7866b90.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "d42dc95dac7b63487440ee8ec6acc7f65dd513460fd86680938267b27fcab401", "generation_seconds_batch": 15.412775260047056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21309423446655273, "target_z": -0.9570312138693761, "ce": 6.260893821716309, "genuineness": 4.331617832183838, "identity": 0.4635213017463684, "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_RANT_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "f266b64ad0381f89296ebaca319d095190d66691033efdb22797b8cd9d818125", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5c3b642525eb1818dced8e851726ef4488cecbf91da52b60cbefb8a368d6eb7d", "generation_seconds_batch": 22.184475932037458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9407780766487122, "target_z": -0.386718748175221, "ce": 6.308549880981445, "genuineness": 4.3127899169921875, "identity": 0.6576255559921265, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f266b64ad0381f89296ebaca319d095190d66691033efdb22797b8cd9d818125.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "f32fc3c9685cb50cdb884f1f723a9f4ba3496a20151df7c2e93a62ae05d22b2d", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "ea28d922b3a6cbf06035ea016d8d2c36ff112ac7b6b6b1cd277ad9142e556450", "generation_seconds_batch": 21.218388702953234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6566823124885559, "target_z": -0.6093749985401768, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.6914960145950317, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f32fc3c9685cb50cdb884f1f723a9f4ba3496a20151df7c2e93a62ae05d22b2d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "f39578c78513f5157710840a16f1834151c2c01fdd8d8b3f08b658e5c2836256", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "750dcb5b495fef8955a1e47a51ee294df671dbaa792e165ddfd5d69d689908d7", "generation_seconds_batch": 21.50867951498367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3401448726654053, "target_z": -1.3906249547454812, "ce": 6.09409761428833, "genuineness": 4.482243061065674, "identity": 0.44365209341049194, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f39578c78513f5157710840a16f1834151c2c01fdd8d8b3f08b658e5c2836256.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": "f3bc34569691f5c7c5a9d621911d539da86bc2ac709e9a7f768076c8ee9b3c1c", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2767280dbe58831de7d61978bfd4c4326d049625ec2f0d0317545f56ae6461d8", "generation_seconds_batch": 20.951573069090955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9824625253677368, "target_z": 0.4296874875915029, "ce": 6.1020402908325195, "genuineness": 3.9801597595214844, "identity": 0.5539014935493469, "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_RANT_reg/f3bc34569691f5c7c5a9d621911d539da86bc2ac709e9a7f768076c8ee9b3c1c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "f45cdc1ef5db88bf4aa29af119ac4bf3c6b46bb5398b48baed3b28a6df82e018", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5a7913c0898cfd3835f44acdd4df8f6a77fda5606a9ce49f5a5ad168ac150ec4", "generation_seconds_batch": 17.502991855959408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.007383346557617, "target_z": 0.44921884744319796, "ce": 6.046441555023193, "genuineness": 3.7667744159698486, "identity": 0.4383939206600189, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f45cdc1ef5db88bf4aa29af119ac4bf3c6b46bb5398b48baed3b28a6df82e018.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": "f47e34d2b9696a24176b4bb428bc5838e991cfcff9094dc76211a4051911e8f6", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa60914fd226cb28c8676211bd58beceb89fd263a4511652addeb54bf9a367d1", "generation_seconds_batch": 20.938480478012934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7599878311157227, "target_z": 1.0390625328460217, "ce": 5.804190158843994, "genuineness": 3.5690791606903076, "identity": 0.4997410774230957, "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_RANT_reg/f47e34d2b9696a24176b4bb428bc5838e991cfcff9094dc76211a4051911e8f6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "f503b75f377d87146a23a4067d38653e2a807c7e4e86b0f0a68784df3c67397b", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "d2891826bcc6db373424ef93d0890386ad7fca209a867938c68636b5fd23aed0", "generation_seconds_batch": 23.334325533010997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.305069923400879, "target_z": 2.249999929928487, "ce": 5.657250881195068, "genuineness": 3.6663577556610107, "identity": 0.3781607747077942, "hypothesis": "Die Tür ist noch offen! Ich habe deinen Mantel auf den Stuhl gelegt! Sag mir, wenn du bereit bist!", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f503b75f377d87146a23a4067d38653e2a807c7e4e86b0f0a68784df3c67397b.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "f5e6ee6202e7658841f76440fb63a6d11d77a86e3bd55a0f6ce2cc6672a293c9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ccb62d48f0d9416abbe6b8adc1f655cda6125c7c7b8cab5c8f72dc2946b9481a", "generation_seconds_batch": 20.412216482916847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.18065261840820312, "target_z": -1.2656250364955797, "ce": 6.284721851348877, "genuineness": 4.406930446624756, "identity": 0.4577872157096863, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f5e6ee6202e7658841f76440fb63a6d11d77a86e3bd55a0f6ce2cc6672a293c9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "f677d2e5223d7fb222f59c3c4492b417103e6433c6c36fddb6fb69a2406a78c3", "bucket": "vn_S_RANT_reg", "phase": "wording", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9bb7bbcaf202abe1b180f5525aa26756c8adfaf8b51062b1c780ec20e3e22743", "generation_seconds_batch": 18.33736811601557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4061717987060547, "target_z": -0.02197265136871621, "ce": 6.252951145172119, "genuineness": 4.3504462242126465, "identity": 0.6746726036071777, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f677d2e5223d7fb222f59c3c4492b417103e6433c6c36fddb6fb69a2406a78c3.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "f758346b0409aa59429f8365d5b0ef7221206fe1836caad27d3108f5b2f29e7f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "320a6f3a6e98a442b79f8dc5b09102b8c0f1a882f696a43f7f87b3b0e1176b1e", "generation_seconds_batch": 21.8986858939752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6870369911193848, "target_z": 1.7656248963525536, "ce": 5.3514580726623535, "genuineness": 3.3243136405944824, "identity": 0.5306358337402344, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f758346b0409aa59429f8365d5b0ef7221206fe1836caad27d3108f5b2f29e7f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "f786d916b7e098523565ba3949b69cb584b1a554fdbb042af77a3c1f8f737e91", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0fb6406373c26c957836f3b5a8cc1bdca5c0993f72940e55d86f90bb93684a31", "generation_seconds_batch": 21.18242799595464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.46728515625, "target_z": -0.7578124832120333, "ce": 5.855817794799805, "genuineness": 4.030367851257324, "identity": 0.3770548105239868, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f786d916b7e098523565ba3949b69cb584b1a554fdbb042af77a3c1f8f737e91.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "f7a1b6e4926e540e059174b6b67743be4830960e77bc35379804fccbdb1d6e35", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bc678a57820ecdb847e660e2c790d1ca50b1beca2ac36deb659326dbe673c8b1", "generation_seconds_batch": 18.447684631915763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6385571956634521, "target_z": 0.1601562912400051, "ce": 6.109982967376709, "genuineness": 3.9425034523010254, "identity": 0.5517874956130981, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f7a1b6e4926e540e059174b6b67743be4830960e77bc35379804fccbdb1d6e35.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "2d5db99b9cdb3ac29e495f191a1c3b23d0e39d1d68dac257fa672bd7c9f00c0e", "generation_seconds_batch": 14.729028196074069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.20058906078338623, "target_z": -1.2812499562053044, "ce": 6.213237762451172, "genuineness": 4.394378662109375, "identity": 0.6802031993865967, "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_RANT_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "f7d8e0b773271abc93bb956203d0b2c6c186271699225ab0b60a5a55ef4b5309", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "292546d2b08a5a019ec48e68188579f08be7eec8bfddd3b7981a4d1887d6d41b", "generation_seconds_batch": 21.827090765931644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8197975158691406, "target_z": 1.0859375722612479, "ce": 5.863759994506836, "genuineness": 3.873466968536377, "identity": 0.578981339931488, "hypothesis": "La porte 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_RANT_reg/f7d8e0b773271abc93bb956203d0b2c6c186271699225ab0b60a5a55ef4b5309.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "f820163efc690e819a60d0527b3888728f25cb2e4e3f2492b1f2c0a721ac8850", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "436a4a656571772cdd114e8aa2c922d8b997de16dcc5b5da7eb3043609c247a9", "generation_seconds_batch": 22.83529062604066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.092113494873047, "target_z": 0.5156249664240666, "ce": 5.95112943649292, "genuineness": 4.13078498840332, "identity": 0.22719421982765198, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f820163efc690e819a60d0527b3888728f25cb2e4e3f2492b1f2c0a721ac8850.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "f82e8554227cd43cc1b47ff03fda20bf721fa788c8e055fdde5ce3e5a4f3dc62", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8b29a850d66eaa07089d77564bb103d053f764bef3896c9ffe2b6982d2daf709", "generation_seconds_batch": 21.5575936390087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.949384927749634, "target_z": 1.1874999708035363, "ce": 5.446769714355469, "genuineness": 3.6914620399475098, "identity": 0.6992684602737427, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f82e8554227cd43cc1b47ff03fda20bf721fa788c8e055fdde5ce3e5a4f3dc62.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "f84e6c25e4eefb9c0ddf04b103196c5d1399a1ba518c1416644249775e634317", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "140b8183de9c7a36095a37f0382f3730cbf678c2d7ab739a720a50f8fee18f8d", "generation_seconds_batch": 15.112451046006754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4672274589538574, "target_z": 0.025878904653318387, "ce": 6.229123115539551, "genuineness": 4.494794845581055, "identity": 0.7144985198974609, "hypothesis": "J'ai trouvé la 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_RANT_reg/f84e6c25e4eefb9c0ddf04b103196c5d1399a1ba518c1416644249775e634317.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_RANT_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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": "baa9bee76bdaa4efb594c1157b8bc5e071c817a2961bcbb26836de08e0a278c1", "generation_seconds_batch": 14.721436911961064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.19062089920043945, "target_z": -1.2734375430647842, "ce": 6.205295085906982, "genuineness": 4.375550270080566, "identity": 0.6437132358551025, "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_RANT_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "f88d246adeb02c4dc48ecf37b3a4268804fe3e53103311be6972a5b4f14f4504", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "68684a6bbca8fc495968af675cd8edbf5e45dd05ad8781c4c6a3df1b9e47b118", "generation_seconds_batch": 24.44391175196506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9163074493408203, "target_z": 1.9453125474442536, "ce": 5.990842819213867, "genuineness": 3.7542223930358887, "identity": 0.6729373335838318, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f88d246adeb02c4dc48ecf37b3a4268804fe3e53103311be6972a5b4f14f4504.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-0", "key": "f98cfbafefcea3d3d42287ae764fb4c6504db09f811527872860ecd80ceaebd6", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "037d758f5e7a82cfcaa3738c71e09bbb9507c5ad38b6f4297a85629407254318", "generation_seconds_batch": 21.443942278972827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6403684616088867, "target_z": 0.9453124540155695, "ce": 6.070269584655762, "genuineness": 4.036643981933594, "identity": 0.5790622234344482, "hypothesis": "La porte 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_RANT_reg/f98cfbafefcea3d3d42287ae764fb4c6504db09f811527872860ecd80ceaebd6.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-0", "key": "f9cdb38b61e1cff691a9436d8f008069656d39b0176ec8f73516687a95840c76", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "841d69793b93e4f1361d92a6f90a76310b7af16ee234c9fbee7a26fdc9515866", "generation_seconds_batch": 21.312847984954715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5869044065475464, "target_z": -0.6640624978102652, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.46039366722106934, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/f9cdb38b61e1cff691a9436d8f008069656d39b0176ec8f73516687a95840c76.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": "f9faa01ab879be453f94337e36a8ea9fbb733d2216330c2131271df8426ca8b7", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9456712834dd0d13cbfbebcf5a597cd5f7587f1ff48b253fa78e7304fdf9a7fa", "generation_seconds_batch": 21.039392792037688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.407925605773926, "target_z": 1.5468750861295681, "ce": 5.879645347595215, "genuineness": 3.4969046115875244, "identity": 0.49074602127075195, "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_RANT_reg/f9faa01ab879be453f94337e36a8ea9fbb733d2216330c2131271df8426ca8b7.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "fa4b6622b4e4bfd80a7df246f3ef2bbc1c8a2f29ba090cbca7f4bfcaa4e3dbf0", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "774b65f53b19092a35f8d94147386927e2febbc149daa9e40ebc543d85c0177b", "generation_seconds_batch": 15.979287159978412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627227306365967, "target_z": 1.7187498569373274, "ce": 5.720792293548584, "genuineness": 3.7793264389038086, "identity": 0.25393998622894287, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fa4b6622b4e4bfd80a7df246f3ef2bbc1c8a2f29ba090cbca7f4bfcaa4e3dbf0.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "fa9c071762cd4f4bcf11fa8aa4149de2b6adf8c7692c670449e484686bbf69bf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f5116e01711b9ef7d8cc0218f985abd5b239b7073cd94066193eb23d72699224", "generation_seconds_batch": 21.340937420958653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9681907892227173, "target_z": -0.3652344018242511, "ce": 5.974957466125488, "genuineness": 4.118232727050781, "identity": 0.694676399230957, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fa9c071762cd4f4bcf11fa8aa4149de2b6adf8c7692c670449e484686bbf69bf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "fab96f4443ba1e97d1b43ed51d74ee33a92b4e2fc198876901e76658dffe9255", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "eb6d46e69dbcb48cb3647693fcc043d9e640662ff63ffbfef5cce432fc8358dd", "generation_seconds_batch": 24.843240538961254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028681159019470215, "target_z": -1.1015624919709726, "ce": 6.046441555023193, "genuineness": 4.419482707977295, "identity": 0.6579692959785461, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fab96f4443ba1e97d1b43ed51d74ee33a92b4e2fc198876901e76658dffe9255.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-es-1", "key": "fb1f005315c6eacc33f5d0a73d99af830ad5f5ec093fb52d973f3c6358dc7e06", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "618807a63f4c26d8e53e68d58a369318f2fc0065e01b762dc1ed9eb9bc0c0317", "generation_seconds_batch": 23.784673233982176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5338902473449707, "target_z": 0.07812497226335939, "ce": 6.149696350097656, "genuineness": 4.086852550506592, "identity": 0.6968547105789185, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fb1f005315c6eacc33f5d0a73d99af830ad5f5ec093fb52d973f3c6358dc7e06.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "fb3c5e86417cd14a2146b43ac7d2fe6bebe7732fada704a15a0ab51a23c063bf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7492d404d80716943ceb3fd3854eb51676c6b6c1ae858700682f06966b7c4364", "generation_seconds_batch": 20.044148677028716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6317616105079651, "target_z": -0.6289062649631877, "ce": 6.181467056274414, "genuineness": 4.093128204345703, "identity": 0.651747465133667, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fb3c5e86417cd14a2146b43ac7d2fe6bebe7732fada704a15a0ab51a23c063bf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-0", "key": "fb43be2d93460e4b4c36eb5e81f500283415a6c794454fdf57e7164b7a6d8a99", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a80e09e4531276ed60ff8d9bfbc959f3163e9e55dbac91e2376beee8072e329d", "generation_seconds_batch": 15.883124582003802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4260506629943848, "target_z": 0.7773437496350442, "ce": 6.173524379730225, "genuineness": 3.8169827461242676, "identity": 0.45640242099761963, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fb43be2d93460e4b4c36eb5e81f500283415a6c794454fdf57e7164b7a6d8a99.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-0", "key": "fb4e3c6dd9c3cbf43d64832745355b4a533d2bb1d5ebc8c0af6a8c92d9e7ca42", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "473237a7a7f3f09bcf015109247a76c58da8ad15ce592efe64f3a16dd5f81235", "generation_seconds_batch": 21.920783741981722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0223355293273926, "target_z": 0.46093742043963615, "ce": 6.09409761428833, "genuineness": 4.080576419830322, "identity": 0.6741204261779785, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fb4e3c6dd9c3cbf43d64832745355b4a533d2bb1d5ebc8c0af6a8c92d9e7ca42.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": "fb6e45a685f240f219ab279b259c1ccfdd3376e4b51c44e755cc76bc9ce3cd5e", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28b10f9072bec60743d26e098223a310269af38fc5b304c7aadd9a8fe98b5e7c", "generation_seconds_batch": 22.73466582398396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.417893886566162, "target_z": 1.5546875926987727, "ce": 5.959072113037109, "genuineness": 3.393350124359131, "identity": 0.614316463470459, "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_RANT_reg/fb6e45a685f240f219ab279b259c1ccfdd3376e4b51c44e755cc76bc9ce3cd5e.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-de-1", "key": "fb954b6be592b19a6860d2c68261016e118df78c5de9fc0fbb4cf0a95d80aba9", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ec87da439acd8ca685f47edf4939fd6c9df3d6dbc8454f01440c9f047a54ca6b", "generation_seconds_batch": 17.502991855959408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5538268089294434, "target_z": 0.0937499854017681, "ce": 6.213237762451172, "genuineness": 3.835810661315918, "identity": 0.541471540927887, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fb954b6be592b19a6860d2c68261016e118df78c5de9fc0fbb4cf0a95d80aba9.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-en-1", "key": "fba30f7f39ec3c25b3fb7d72faf067e43c8d3aa4cb73d69297fa6a6f6d25eedf", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b923ef997c8bc2eb8a6d01783957afa669cf6a2cdf62936bff70395024c022a3", "generation_seconds_batch": 19.820723601034842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6616664528846741, "target_z": -0.6054687452555746, "ce": 6.054384231567383, "genuineness": 4.099404335021973, "identity": 0.3119822144508362, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fba30f7f39ec3c25b3fb7d72faf067e43c8d3aa4cb73d69297fa6a6f6d25eedf.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "fbb39c118dfdf6cab6f0fe2a44f6f6d9e477f2432fe8f1a0c6dd0e2839426557", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0de64c87e4054da6aa3efbcef8264fc7ee7e6d7c9c59c209442cc9817a9716b1", "generation_seconds_batch": 17.06083371094428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2783379554748535, "target_z": 1.4453125007299117, "ce": 5.736677646636963, "genuineness": 3.575355052947998, "identity": 0.7534676194190979, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fbb39c118dfdf6cab6f0fe2a44f6f6d9e477f2432fe8f1a0c6dd0e2839426557.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "fbb75c31513d075e24b7e269cc512659ee7629133c318a35790bd9f96796e42f", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e97cc36e0f87af0f3d0c6e8ca69585d070ac80c7712665f1e118b7a6c221e6b6", "generation_seconds_batch": 25.13324150198605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3979573249816895, "target_z": 1.5390625795603639, "ce": 5.6532793045043945, "genuineness": 3.5565271377563477, "identity": 0.569061815738678, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fbb75c31513d075e24b7e269cc512659ee7629133c318a35790bd9f96796e42f.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-fr-1", "key": "fd61495dcdc24bc5b235a0769e49e5a96a85bb13a001dc6a8723425388600c39", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3eec3abc3a75826333806d9d4be29e1a8c44db374f2af7a55ebc726ce4726bb6", "generation_seconds_batch": 22.93579800100997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.328179359436035, "target_z": 1.4843750335759334, "ce": 5.7684478759765625, "genuineness": 3.722842216491699, "identity": 0.47018975019454956, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/fd61495dcdc24bc5b235a0769e49e5a96a85bb13a001dc6a8723425388600c39.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": "ff454a1fd3b5a99f4bf3ea60b0ff4f276216913cb056906058e42ad68f4da776", "bucket": "vn_S_RANT_reg", "phase": "heldout", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b52544722479833677810d613903b3371cdb23ba4c4bb6e16736b15b7db38784", "generation_seconds_batch": 15.844272442976944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7069735527038574, "target_z": 1.7812499094909622, "ce": 5.502368927001953, "genuineness": 3.4435582160949707, "identity": 0.4923746883869171, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_RANT_reg/ff454a1fd3b5a99f4bf3ea60b0ff4f276216913cb056906058e42ad68f4da776.mp3" }, { "id": "vn_S_RANT_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_RANT_reg-nl-1", "key": "ffa49ee12839bb45b44106fa082499c22257819f9ef4a04242aa56f24187f4d3", "bucket": "vn_S_RANT_reg", "phase": "grid", "candidate": "manual-9663bdd37f37", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(A forceful sustained rant; emphatic stresses, rising insistence and frustrated breath without losing intelligibility. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "931bc4d2eef4397ed2190f6c207d279afbeb8ffd524df704656c3560ea319c64", "generation_seconds_batch": 18.96602002205327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0291311740875244, "target_z": 1.250000023357171, "ce": 6.125868320465088, "genuineness": 3.550251007080078, "identity": 0.4906938672065735, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen het 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_RANT_reg/ffa49ee12839bb45b44106fa082499c22257819f9ef4a04242aa56f24187f4d3.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "134e3202ae30fc737acb2c971e6018ba95244264d93b7fbbcfa6250b00b9bd84", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3875389099121094, "target_z": -0.8203125357656682, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.4502226710319519, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/3a52cf60914f551d7d623faf.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "53504b35ddd1eab9ab01cd181d469e31162e1e05519e3a1c8aff3b0af25ab6bf", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.021160244941711426, "target_z": -1.1406250248169942, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.5993266105651855, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/ce7cf61a0792fafc15cac870.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "af72c5cbf448c0a6c1f82a33f0860a0f7314dae9e5e0919bff6ae7dde140f6dc", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19814181327819824, "target_z": -0.9687499737231826, "ce": 6.221180438995361, "genuineness": 4.469690799713135, "identity": 0.6910701990127563, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/e89aa4cf6aef3317f80daa38.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "958e256575489ee111b44bf6b6885394962fb90b37c5946597ec8b83690e850a", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018712878227233887, "target_z": -1.1093749985401768, "ce": 6.109982967376709, "genuineness": 4.7081804275512695, "identity": 0.6806373596191406, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/13ea9d04a8b3c15989fe900a.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "335fba11e8c84c9add3033a5345108ec114c7a0f83b93545c56d1eb4f2de8a8c", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3476659059524536, "target_z": -0.8515624686138015, "ce": 6.086154937744141, "genuineness": 4.20609712600708, "identity": 0.4767787456512451, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/09a5fa5b916d113d54d08679.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d192f85b19fe5110db67dcdf36ee112731c69850565672eb8491077a2db30d80", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22804665565490723, "target_z": -0.9453124540155695, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.5614414215087891, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/d1284ef4c4370be7cc14d251.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "b9a52f3c5ed87f259ae51db4e7f6fc02f2cb1f8d76e4f32af32bf7ebf34b761d", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6915712356567383, "target_z": -0.5820312722623037, "ce": 6.197352409362793, "genuineness": 4.15588903427124, "identity": 0.44962647557258606, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/548d25a1942936e450f08ee1.mp3" }, { "key": "original-vn_S_RANT_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "5921c69b89ebd5be656fe20a03911d293b25b7be25a8e5e7bdc9a9650e2ff2ae", "task_id": "vn_S_RANT_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39252305030822754, "target_z": -0.816406282481066, "ce": 6.221180438995361, "genuineness": 4.369274139404297, "identity": 0.5815708637237549, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/63af74c8d9a54135ea80349f.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "7dcffae5ef5f4177fa8caa7d3b0ec3ee5ce7b66f00e170eaf484a106af4903f8", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3401448726654053, "target_z": -1.3906249547454812, "ce": 6.395919322967529, "genuineness": 4.093128204345703, "identity": 0.22401022911071777, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/28d418ef5cca325d965320e7.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "97b8278171afa8c78e4cf89e32b451b00fb45803d5178c646ebd67f3530cc198", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2180783748626709, "target_z": -0.9531249605847739, "ce": 6.38003396987915, "genuineness": 4.124508857727051, "identity": 0.45168831944465637, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/3be464801a7b6a4b048edbca.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "5409fba2b3b2351d981710be67195627e2e0df6ba709c186ddf13f911497f60c", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028681159019470215, "target_z": -1.1015624919709726, "ce": 6.340320587158203, "genuineness": 4.231201171875, "identity": 0.4428517818450928, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/c8d47723ecaf3c763bd15ec8.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "83746c5345a37a0d1893e2d2225f5c939ea295eb42b5015a4ac3ba330d4c89c5", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.2603987455368042, "target_z": -1.3281249956205303, "ce": 6.332377910614014, "genuineness": 4.243753433227539, "identity": 0.44162920117378235, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/de/cbc0ac36a4e13965bcf3eb4b.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d4697ed11d20dc1a02dbf58c8b51c81597dda5cf51da8ef39d3c2e5a9142cae6", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.2205256223678589, "target_z": -1.296874969343713, "ce": 6.403861999511719, "genuineness": 4.149612903594971, "identity": 0.5656195878982544, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/78c94b1c54751a20c70a4280.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d1a18326eab9cc9f477e610bee6812cc9a2996df153d09d3c9cf4b619e95002b", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.2603987455368042, "target_z": -1.3281249956205303, "ce": 6.324435234069824, "genuineness": 4.256305694580078, "identity": 0.5265320539474487, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/0525955f842d9a0f7275ee74.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "bedd8cc441500fbaef1fcbb3f3d0b605a47422e9c99cff81e93da47a76527b8e", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0012236833572387695, "target_z": -1.1250000116785854, "ce": 6.332377910614014, "genuineness": 4.18099308013916, "identity": 0.45462605357170105, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/893d4a255ec8107c866b8351.mp3" }, { "key": "original-vn_S_RANT_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "80651c0d309d7ae675631ff71c076d2ad6ebb7058b0a91f5992267eb3a0d39ad", "task_id": "vn_S_RANT_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05106508731842041, "target_z": -1.1640625445246073, "ce": 6.348263263702393, "genuineness": 4.419482707977295, "identity": 0.5849694013595581, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/de/c3659e22fa32018015bdeff6.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "316f71232d5d7dd270439552118f7d86c93228781fbb9592e32b61731d01157f", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.008744597434997559, "target_z": -1.1171875051093811, "ce": 6.157639026641846, "genuineness": 3.798154592514038, "identity": 0.4272165298461914, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/e91b3f9dedd9b01449b30571.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "782de48b9276e3ade8356e0ca5d64b25df697aecdcb35b8b7496906322636ae3", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4971899390220642, "target_z": -0.7343750102187623, "ce": 6.3641486167907715, "genuineness": 4.1998209953308105, "identity": 0.3124083876609802, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/025e3f31d0a72f8afc6abafa.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "3246b303caa6ff9cd17a83c2aa3d169534d0a4584a9eebc7ca55c140223f679a", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5569995641708374, "target_z": -0.6875000175178783, "ce": 6.221180438995361, "genuineness": 4.042920112609863, "identity": 0.31297093629837036, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/en/9e4f04eec38c27084dbd1052.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "44d31130d58bfeb450dcf02fb28da90d91550dd920cad1083641ccbed2d11aa9", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.267919659614563, "target_z": -0.9140625211674362, "ce": 6.252951145172119, "genuineness": 4.143336772918701, "identity": 0.39968353509902954, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/207bc91b1e742508f22e3e61.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "0c2f974c8362b2a18d49c1092e18d4b3d09dcb048e55c27ae8151fae42149088", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.267919659614563, "target_z": -0.9140625211674362, "ce": 6.24500846862793, "genuineness": 3.8922953605651855, "identity": 0.16769763827323914, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/307c3762edbe0d9b946eaa09.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "71c39edd9607a7ec1c257e837cb66dc518c0eba6869cd43433b9596a058cf9dd", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6168092489242554, "target_z": -0.6406249781026522, "ce": 6.300607204437256, "genuineness": 4.086852550506592, "identity": 0.11525426805019379, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/ac2cb71e71fc9cdd9d5128da.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "e312daea5d7607be63498349f7dfc0b45fb29d92fd6487106b96390b288ae76b", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09845900535583496, "target_z": -1.0468750394152262, "ce": 6.1894097328186035, "genuineness": 4.137060642242432, "identity": 0.179977685213089, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/25f96f006d5bda105df907c6.mp3" }, { "key": "original-vn_S_RANT_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d847f9ed73bb816aa6c58ed9bef9a8ecd31778c152911712007381d3acffe750", "task_id": "vn_S_RANT_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32772934436798096, "target_z": -0.8671874817522102, "ce": 6.173524379730225, "genuineness": 4.061748504638672, "identity": 0.28942230343818665, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/00a1f1a59b46be1d0c17da2b.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "8d052edc9e8fe558dbf122fd3860848cffcea9462d260e6518d9085b2dcfa7d7", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.4597642421722412, "target_z": -1.4843750335759334, "ce": 6.38797664642334, "genuineness": 4.494794845581055, "identity": 0.26738256216049194, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/5dc0bd6d9d989e9a2a744b6c.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "89cb7550b6565f7ac9f3c8a301c3e5fe2d85f93775552abb3244e69014930fa8", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.43982768058776855, "target_z": -1.4687500204375246, "ce": 6.443574905395508, "genuineness": 4.557555198669434, "identity": 0.5412545204162598, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/976e10a35ad33c4005fc6a62.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "46c94e0d901370ff64b116d03735f2da60f6c6520de46a2fe053bbf95a0b3156", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.24046218395233154, "target_z": -1.3124999824821217, "ce": 6.356205940246582, "genuineness": 4.356721878051758, "identity": 0.6107213497161865, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/c611f30e93f9ad378bce384e.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "73d15ecbad345adccd3027da686bc7c7817ec0cb686afb898c1e9b6174237d35", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3899862766265869, "target_z": -1.4296874875915029, "ce": 6.340320587158203, "genuineness": 4.519899368286133, "identity": 0.5053507089614868, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/829feab2cba4f81e98c4693c.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "4d4e6d19f943f36a6967c2ddb3f5efaea96319dcd7e02f9997217e4ce3003121", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.24046218395233154, "target_z": -1.3124999824821217, "ce": 6.356205940246582, "genuineness": 4.281409740447998, "identity": 0.3896414041519165, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/b38620ccf1df3cb97ceed440.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "5a9034c14702ed30baa0e5c8d18ae0056bab7bcf783d4a8a22e6ad19be9c4c17", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3401448726654053, "target_z": -1.3906249547454812, "ce": 6.252951145172119, "genuineness": 4.2374773025512695, "identity": 0.4108528792858124, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/0bcc5ec9e9a0c4737be33f7f.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "859fc5978104183e511196fe4fd0d2da3cd16463ae8c33883e565cf1386bc3f9", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3401448726654053, "target_z": -1.3906249547454812, "ce": 6.3641486167907715, "genuineness": 4.356721878051758, "identity": 0.2680177688598633, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/8938af1dddbb8612f7ff34f6.mp3" }, { "key": "original-vn_S_RANT_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "1971f885760bd52bb99d334c76e21392feeeed8620180baa174e48e7a4cc5eb8", "task_id": "vn_S_RANT_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.4198911190032959, "target_z": -1.453125007299116, "ce": 6.435632705688477, "genuineness": 4.3881025314331055, "identity": 0.4032140374183655, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/en/5bdf85a2a8434a04c7e4526c.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "efcecac15ed1dacdd26fbd696889527d7df2f2d28116393b8df0bf1cd399c76b", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.07100152969360352, "target_z": -1.1796874642343318, "ce": 5.434855937957764, "genuineness": 3.9550557136535645, "identity": 0.6113851070404053, "hypothesis": "La puerta sigue abierta, ha 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_RANT_reg/es/6bbe821f829f60d30a907b25.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "f93736935c75d6d4b3bf381bf71b0d8cd811617cd71daa9704f9e4f5799360dc", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.2603987455368042, "target_z": -1.3281249956205303, "ce": 6.149696350097656, "genuineness": 4.168440818786621, "identity": 0.47754207253456116, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/7bfc718a8926ecad3cd654bc.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "49633eb66b3ebef2de7a5bc5b7398646355fff724a2b3f950f341a74e8f26806", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06103324890136719, "target_z": -1.1718749576651275, "ce": 6.173524379730225, "genuineness": 4.193545341491699, "identity": 0.49551042914390564, "hypothesis": "La puerta sigue abierta. His 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_RANT_reg/es/4010af3239d7ff198a3b6b70.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "1af0d9d2fa215adecf020acbade84f52e1cd6120402026d43f2973c08c9bb540", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.37004971504211426, "target_z": -1.414062474453094, "ce": 5.967014789581299, "genuineness": 3.5722172260284424, "identity": 0.5451236367225647, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/36af8c7bf0d680e4f32e7004.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "90ef174898af46fb0f975382ffb6e5968d5bf55112da1964985f086e3ed6fe84", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3401448726654053, "target_z": -1.3906249547454812, "ce": 6.0305562019348145, "genuineness": 4.005263805389404, "identity": 0.6102416515350342, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/27b38879e046787389d659b5.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "ce0a122a8fdfa576622f5a5093e58a9fa1fb2f4c5704d8fe21e9c47d453a720c", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018712878227233887, "target_z": -1.1093749985401768, "ce": 6.006728172302246, "genuineness": 4.080576419830322, "identity": 0.5722070336341858, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/b7439a4cb38e1d85263faa3b.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "2d66755bcaaeeff258b3a67482696636975b834b9a86cbdd5c73e2b81db3e365", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.32020843029022217, "target_z": -1.3750000350357565, "ce": 6.149696350097656, "genuineness": 4.256305694580078, "identity": 0.3927333950996399, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/803181a8035df91db4e8b08d.mp3" }, { "key": "original-vn_S_RANT_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "5a2f030ed4c3d0ca744e86994700cfdba51568f03b51d6f5589b5d108d1215c4", "task_id": "vn_S_RANT_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.24046218395233154, "target_z": -1.3124999824821217, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.40685564279556274, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/2bf11d74d5858386908570db.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "374cf34d3fd4b9f55060269f895753171795d1a1f60a4915ed582f3d868cb184", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.5594469308853149, "target_z": -1.5625000058392928, "ce": 6.078212261199951, "genuineness": 4.444586753845215, "identity": 0.7388945817947388, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/68d01f349fe6f92fa83b5a51.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "ecadf6cf2a2ad1fb28bab817d2e3b82943652dd0c8eda8f89b6cea842953e1ff", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.6391931772232056, "target_z": -1.6250000583929276, "ce": 6.086154937744141, "genuineness": 4.18099308013916, "identity": 0.3845941424369812, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/b2037762dbe576cf1ad6fe2c.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "db07b52fb7aee672a68ff757309696677589e4769cd6cc7a4887fe91a9b8d6d5", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.18065261840820312, "target_z": -1.2656250364955797, "ce": 6.046441555023193, "genuineness": 4.519899368286133, "identity": 0.5842911005020142, "hypothesis": "He encontrado la quejita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/es/b74dc9fd6ea955b592d4fe15.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "e872608d5a55c9ad453359d1ea52dc260ac5c1932571a44344cb8bd0c57b0f9d", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.16071605682373047, "target_z": -1.250000023357171, "ce": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.6830938458442688, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/a1fcf046b4401ca65817881e.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "17eb0af0cd739f9f3154f003dba2d41d9a62277395db3079077626e454782bd8", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.08096981048583984, "target_z": -1.1874999708035363, "ce": 6.109982967376709, "genuineness": 4.381826400756836, "identity": 0.6515319347381592, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/48a514dee11f72e9883e9e27.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "0e9af24950e37ff341e36241fe5a7fd732497ee2d52a2c946ad1d9c11d323af0", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12836384773254395, "target_z": -1.023437519707613, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.6334404945373535, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/2ead84179bebe301f39db391.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "ee21244418d64b8be0de0840677b6ff3dcb15e258c0c9de56bdc23eeba17c439", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.21055734157562256, "target_z": -1.2890624627745086, "ce": 6.213237762451172, "genuineness": 4.482243061065674, "identity": 0.565312922000885, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/0a2fdd2c7d38cd800e2d65b3.mp3" }, { "key": "original-vn_S_RANT_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "bba165be5f684c6cedf81b262ddc8cddac0642b8da6322ce06b448a2abfc4f8c", "task_id": "vn_S_RANT_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.43982768058776855, "target_z": -1.4687500204375246, "ce": 6.133810997009277, "genuineness": 4.7332844734191895, "identity": 0.5793610215187073, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/es/cc264b53887a96ade0403fe2.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "5035884b8c99df921622e6c525f9796ca69862a3542f83132691d990b52f82a8", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27290380001068115, "target_z": -0.9101562678828341, "ce": 6.23706579208374, "genuineness": 4.231201171875, "identity": 0.7719144821166992, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/b20944613e6eb6856d86fb9b.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "47bef0d040bfd8a1704f2411f1231568ee46efeef961533ceff2cf0032ff6079", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09845900535583496, "target_z": -1.0468750394152262, "ce": 6.197352409362793, "genuineness": 4.394378662109375, "identity": 0.4947175979614258, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/c527eb11ef2fe2f61b3bfb0a.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "0bde4281d11808664639d3af22e34f221c091608c6ab20aee0a63445d52f3b05", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.48966896533966064, "target_z": -1.5078124598548623, "ce": 6.022613525390625, "genuineness": 4.149612903594971, "identity": 0.7808941006660461, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/d090299f410970caeba3741a.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "b910d0070a439d9816d05af193b3d2017658971abcc06fc873273225686e9953", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.39995455741882324, "target_z": -1.4374999941607072, "ce": 6.181467056274414, "genuineness": 4.369274139404297, "identity": 0.8207980990409851, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/ca541f33499c63d94e9c343f.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "e36696776842da455378e0ae6c2f81691ab1c5f44323035cec4c24838a48f181", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08849072456359863, "target_z": -1.0546875459844305, "ce": 6.23706579208374, "genuineness": 4.645420074462891, "identity": 0.7315159440040588, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/47aa75ba4b21bbaedfe75a32.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d718d351d640f84384283d1865d21c8d471103fe3178db6148db44a36320e429", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19814181327819824, "target_z": -0.9687499737231826, "ce": 6.022613525390625, "genuineness": 4.557555198669434, "identity": 0.6117023229598999, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/675da0bfa11ea044bc1a3415.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "11e54d5bb38b295d78f59caba6f31ded20b3814229a66ac2a37cd8e6a82705d1", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15826869010925293, "target_z": -1.0, "ce": 6.1020402908325195, "genuineness": 4.344170093536377, "identity": 0.6966162919998169, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/67fe40133545e81741e7cd2b.mp3" }, { "key": "original-vn_S_RANT_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "e489a9f10b6c55395908a4738d8303c2f69c8d3b7bfdfc5acc3351854fd09eaa", "task_id": "vn_S_RANT_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19315767288208008, "target_z": -0.9726562270077848, "ce": 6.324435234069824, "genuineness": 4.406930446624756, "identity": 0.6324079632759094, "hypothesis": "La porte est encore ouverte. Encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/fr/99b636aa2a19f851c0d6d14a.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "c9484e2ae848a03dd4b6f60d352e44a375ce64993f7b0e7a84ff9a1a279d1e87", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.32020843029022217, "target_z": -1.3750000350357565, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.7154766321182251, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/def055074ebcdd4b5a757d20.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "af4f2c1af83e16aaecb5972b792f43d2dc2b1bc2429ffef3ab5b7bef7aa05b73", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.3800179958343506, "target_z": -1.4218749810222986, "ce": 6.109982967376709, "genuineness": 4.406930446624756, "identity": 0.6593051552772522, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/13acf41d287f1e2d8718afa8.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "758bc325391d21fbe84d24ebac645edd778c67541d0ad1810c51cd46b5d5f9de", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.25043046474456787, "target_z": -1.320312489051326, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.5211820006370544, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/a2d1a8ae1b06ea5f162b1635.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "60523ebec2556d572a96675b8aacd7041f640ceefc9b17a7ac5ba46ec920eb29", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.18065261840820312, "target_z": -1.2656250364955797, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.8007286190986633, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/fd6af9921fc441061194cf3c.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "88237053edb8cb93fcb3c86e1d83b1b8101aa3c232ef4ce136632e2ea686d41d", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.20058906078338623, "target_z": -1.2812499562053044, "ce": 6.252951145172119, "genuineness": 4.6956281661987305, "identity": 0.8684443831443787, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/df7e26231efd4742507b8808.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "c4ee47187d7367542265f634487eec15f193d0641a04dca1de8e16d17947cc88", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.12084293365478516, "target_z": -1.2187499970803537, "ce": 6.252951145172119, "genuineness": 4.432034492492676, "identity": 0.6869447231292725, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/7238a2a5fc4303a49b0d8ec3.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "1e4f739dca4bb76dcce9793bf563a7a6f0edf7980865c1ed5beed9a06fda514a", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.20058906078338623, "target_z": -1.2812499562053044, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.7759860754013062, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/6439dc867794b256834bc9da.mp3" }, { "key": "original-vn_S_RANT_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "ad56e00d122789d97ce0f600ee8647fafe3e3884668743e0b833c84a9a2da5fd", "task_id": "vn_S_RANT_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11839556694030762, "target_z": -1.0312500262768174, "ce": 6.181467056274414, "genuineness": 4.632867813110352, "identity": 0.8270875215530396, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/fr/61151572e0e09d18abf2a3a4.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "3345a31db1463a52d6b66cf1bf46453e5ec9d2bf0b63eba7e3babf76ab30f310", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5899617671966553, "target_z": 0.1220702750007918, "ce": 6.1894097328186035, "genuineness": 4.042920112609863, "identity": 0.38304418325424194, "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/audio/vn_S_RANT_reg/nl/b1884120e7f5a8c3ac761931.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "ee16e6d1fa050085f13e20265fd69561cd9b364a8292db6f37b11d82d21b71bb", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9507463574409485, "target_z": -0.37890624160601666, "ce": 6.0146708488464355, "genuineness": 4.231201171875, "identity": 0.6625149250030518, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/8ecdc94f5e7ef1f0e29dea13.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d205e99d5c37c23ab05c5509d003f9378e6de0f17c0495a3b4ff5967d5410533", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7214760780334473, "target_z": -0.5585937525546906, "ce": 6.221180438995361, "genuineness": 4.300237655639648, "identity": 0.6335574388504028, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/nl/aef44b7805e660223f56be79.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "8bc3b18108fe9d08f94d92d7c0cf475306e9262440bc803f738fe6e2014547f2", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6865871548652649, "target_z": -0.5859374788325638, "ce": 6.157639026641846, "genuineness": 4.20609712600708, "identity": 0.5863907933235168, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/nl/9b5191be5e4c8f7b367cff5c.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "593f98f4975add99c0b0f59cde51d617d55f7490c525f9862d54aa15e95d04c7", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6367457509040833, "target_z": -0.6250000116785855, "ce": 6.054384231567383, "genuineness": 4.268857479095459, "identity": 0.7114071249961853, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/1fc72bae0e7c6d920b0ae744.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "d9e084e98fbbe9e80240c6256a2e9c93571a546512c6786c4703ebd5c7286236", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3077927827835083, "target_z": -0.8828124948906189, "ce": 6.157639026641846, "genuineness": 4.256305694580078, "identity": 0.624732494354248, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/5a6f024e9d32cc3c131b0612.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "0f69f8e6f280869d4594c35411548d85263ec96e77bc00ec44da3098f870b667", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.267919659614563, "target_z": -0.9140625211674362, "ce": 6.09409761428833, "genuineness": 4.375550270080566, "identity": 0.6433919668197632, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/8118981b68012d74807c0266.mp3" }, { "key": "original-vn_S_RANT_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "faea53af2dbe63abbab3717f958d8bf2fbfc7b1bf546ba50ad30db82e01403d6", "task_id": "vn_S_RANT_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21309423446655273, "target_z": -0.9570312138693761, "ce": 6.1020402908325195, "genuineness": 4.262581825256348, "identity": 0.7736520767211914, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/772e87c782ccbf3144825fee.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "74aaee8081f71c05fef00d4caffb20aad667aa7ba4407598a45908a7951a4b0d", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47725343704223633, "target_z": -0.7499999766428289, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.5484073162078857, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/7fd093d4d64914530e3c4d3e.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "2f2056766c828c5da5c75cd3d144e28252e7f28cc37f0beaa34568d23e3b94e1", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40249133110046387, "target_z": -0.8085937759118617, "ce": 6.078212261199951, "genuineness": 4.413206577301025, "identity": 0.6955951452255249, "hypothesis": "Ik 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/audio/vn_S_RANT_reg/nl/3a230c5695f19aab9463945e.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "7d32316950b5151cb9cc6640f5ac527b756cefd3b702f28b07fc212bf5776c68", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6865871548652649, "target_z": -0.5859374788325638, "ce": 6.09409761428833, "genuineness": 4.042920112609863, "identity": 0.7010055780410767, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/e3a76b95397f2e7467c516f7.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "b983d858563340a377167a20560706b88827ddda9c13fc33adc63a5fcd8e4bf0", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19814181327819824, "target_z": -0.9687499737231826, "ce": 6.133810997009277, "genuineness": 4.20609712600708, "identity": 0.5832160711288452, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen het er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_RANT_reg/nl/e784533aaa5097e4e1f0ebff.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "c622dd3291e519096edaba9377b36c5a74065250617415ec38c3878d6663c54f", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6367457509040833, "target_z": -0.6250000116785855, "ce": 6.268836498260498, "genuineness": 4.381826400756836, "identity": 0.5915825366973877, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/9548b559924097febcd2e8b5.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "3abb4aff1a641842295860f7bcd6f68062bf9572fb41b5c483aab705cf6efd40", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2834374904632568, "target_z": -0.11816402171618963, "ce": 6.0305562019348145, "genuineness": 4.18099308013916, "identity": 0.6204534769058228, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/0e4e6249b5b809466aca497f.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "9410ed9df4db580379cc74d3e2edf1ee47f12e506e52cf187930c9d1c35f7f4e", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5918885469436646, "target_z": -0.660156244525663, "ce": 5.967014789581299, "genuineness": 4.030367851257324, "identity": 0.5554953217506409, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_RANT_reg/nl/20317aa248fae7afb27937c0.mp3" }, { "key": "original-vn_S_RANT_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_RANT_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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ac78a8fdc59253bf962b5d5288b82e64a0c7917132fb8808b1c79ec8baad6966", "mp3_sha256": "b9ea700055b0f901589a7a7478f70dd1b572ab9bdb0b3202142040ab8cf12981", "task_id": "vn_S_RANT_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Ranting/Angry Style. 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 Ranting/Angry Style. 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 Ranting/Angry Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8460794687271118, "target_z": -0.4609375138683203, "ce": 6.165581703186035, "genuineness": 4.319066047668457, "identity": 0.623521625995636, "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/audio/vn_S_RANT_reg/nl/ccc26bc1cb187ea6e5b9eff7.mp3" } ], "source_standard_audio_unchanged": true }