{ "complete": true, "report": { "bucket": "vn_VOLT_reg_low", "label": "Emotionally steady, stable delivery without abrupt changes", "definition": { "id": "vn_VOLT_reg_low", "category": "bipolar", "head": "vn_VOLT_reg", "pole": "low", "label": "Emotionally steady, stable delivery without abrupt changes", "floor": 1.0, "canonical_dimension_name": "Volatility" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P05", "general": "Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.", "inline": [ "Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.", "Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.", "Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end." ], "recipe": "P05", "cfg": 1.0, "timing_syntax": "canonical", "canonical_labels": true, "lexical_proxy": false, "origin": "new_canonical_manual_extension_requires_new_evidence" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.9839843786193092, "target_raw": 0.9614020705223083, "ce": 6.120705628395081, "genuineness": 4.303375840187073, "identity": 0.5968862056732178, "wer": 0.062283711730306154, "valid_fraction": 0.95, "identity_delta": -0.020505861192941666, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.12265624554959, "target_raw": 0.8409715473651886, "ce": 6.077418088912964, "genuineness": 4.222414863109589, "identity": 0.6104578420519828, "wer": 0.044875085999312, "valid_fraction": 1, "identity_delta": -0.00693422481417656, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.9839843786193092, "target_raw": 0.9614020705223083, "ce": 6.120705628395081, "genuineness": 4.303375840187073, "identity": 0.5968862056732178, "wer": 0.062283711730306154, "valid_fraction": 0.95, "identity_delta": -0.020505861192941666, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.9082031191286761, "target_raw": 1.0272148191928863, "ce": 6.132023930549622, "genuineness": 4.23433928489685, "identity": 0.6347040511667729, "wer": 0.009791666666666667, "valid_fraction": 1, "identity_delta": 0.017311984300613405, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.8527343627479678, "target_raw": 1.0753870368003846, "ce": 6.103033137321472, "genuineness": 4.2829787015914915, "identity": 0.6118389442563057, "wer": 0.01863562091503268, "valid_fraction": 1, "identity_delta": -0.0055531226098537445, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.8164062657104831, "target_raw": 1.1069364190101623, "ce": 6.13738522529602, "genuineness": 4.242184448242187, "identity": 0.6047595009207726, "wer": 0.009791666666666667, "valid_fraction": 1, "identity_delta": -0.012632565945386887, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.7691406147318856, "target_raw": 1.1479845941066742, "ce": 6.207677841186523, "genuineness": 4.261483204364777, "identity": 0.5984492793679237, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.018942787498235702, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.7031249899195533, "target_raw": 1.205316311120987, "ce": 6.215223431587219, "genuineness": 4.291451346874237, "identity": 0.5981840088963508, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.019208057969808578, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.6652343773324438, "target_raw": 1.2382226705551147, "ce": 6.179084229469299, "genuineness": 4.2594434976577755, "identity": 0.5873744778335095, "wer": 0.008596491228070175, "valid_fraction": 1, "identity_delta": -0.030017589032649995, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.680859381836473, "target_raw": 1.2246530294418334, "ce": 6.191792559623718, "genuineness": 4.2801544785499575, "identity": 0.5608158320188522, "wer": 0.008742690058479531, "valid_fraction": 1, "identity_delta": -0.05657623484730721, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.12265624554959, "target_raw": 0.8409715473651886, "ce": 6.077418088912964, "genuineness": 4.222414863109589, "identity": 0.6104578420519828, "wer": 0.044875085999312, "valid_fraction": 1, "identity_delta": -0.00693422481417656, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 2.0355468701474444, "target_raw": 0.9166222751140595, "ce": 6.129641127586365, "genuineness": 4.287999486923217, "identity": 0.621584078669548, "wer": 0.012524509803921568, "valid_fraction": 1, "identity_delta": 0.004192011803388596, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 2.0039062373994416, "target_raw": 0.9441007971763611, "ce": 6.093700575828552, "genuineness": 4.253481280803681, "identity": 0.6312399089336396, "wer": 0.05086278809769522, "valid_fraction": 0.9, "identity_delta": 0.013847842067480087, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.9851562429758591, "target_raw": 0.9603843569755555, "ce": 6.111571502685547, "genuineness": 4.23088755607605, "identity": 0.6661545008420944, "wer": 0.03798546611627107, "valid_fraction": 1, "identity_delta": 0.04876243397593498, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.9699218690750566, "target_raw": 0.9736147522926331, "ce": 5.994615697860718, "genuineness": 4.257560670375824, "identity": 0.6442489773035049, "wer": 0.036466503267973856, "valid_fraction": 0.95, "identity_delta": 0.026856910437345505, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.9343749920643294, "target_raw": 1.0044856786727905, "ce": 6.1500934839248655, "genuineness": 4.186013889312744, "identity": 0.5773999497294426, "wer": 0.009791666666666667, "valid_fraction": 1, "identity_delta": -0.03999211713671684, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.9230468555629687, "target_raw": 1.0143236756324767, "ce": 6.130832552909851, "genuineness": 4.235908389091492, "identity": 0.6285232789814472, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": 0.01113121211528778, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.9187499875603, "target_raw": 1.0180553197860718, "ce": 6.13778235912323, "genuineness": 4.313103652000427, "identity": 0.5985220938920974, "wer": 0.014808866529067767, "valid_fraction": 1, "identity_delta": -0.018869972974061965, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.9011718712198324, "target_raw": 1.0333211541175842, "ce": 6.1433422565460205, "genuineness": 4.23402556180954, "identity": 0.6273912072181702, "wer": 0.01466266769865841, "valid_fraction": 1, "identity_delta": 0.009999140352010726, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.8792968800134136, "target_raw": 1.0523186385631562, "ce": 6.177495718002319, "genuineness": 4.24940185546875, "identity": 0.5825102217495441, "wer": 0.026451023391812864, "valid_fraction": 0.95, "identity_delta": -0.034881845116615295, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.8347656226675562, "target_raw": 1.0909921109676362, "ce": 6.177495718002319, "genuineness": 4.2851752758026125, "identity": 0.605706912279129, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.01168515458703041, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.832421887091174, "target_raw": 1.0930275440216064, "ce": 6.158036136627198, "genuineness": 4.222414886951446, "identity": 0.5960528522729873, "wer": 0.009089912280701754, "valid_fraction": 1, "identity_delta": -0.021339214593172073, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.8007812474798883, "target_raw": 1.1205060720443725, "ce": 6.1411579847335815, "genuineness": 4.251598501205445, "identity": 0.6349502772092819, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.017558210343122483, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.7679687641019013, "target_raw": 1.1490022957324981, "ce": 6.160021829605102, "genuineness": 4.241556870937347, "identity": 0.5947653308510781, "wer": 0.008388157894736842, "valid_fraction": 1, "identity_delta": -0.022626736015081406, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.7523437527345893, "target_raw": 1.162571942806244, "ce": 6.177495718002319, "genuineness": 4.259757339954376, "identity": 0.6305601626634598, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.013168095797300338, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 1.751953128994645, "target_raw": 1.1629111826419831, "ce": 6.177098608016967, "genuineness": 4.2572468519210815, "identity": 0.5975815847516059, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.01981048211455345, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.7398437450133961, "target_raw": 1.1734276592731476, "ce": 6.236271500587463, "genuineness": 4.235908436775207, "identity": 0.5682548850774765, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.04913718178868294, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.7199218930965463, "target_raw": 1.1907289266586303, "ce": 6.2203861474990845, "genuineness": 4.209549009799957, "identity": 0.5770668402314186, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.04032522663474083, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.7113281296380778, "target_raw": 1.1981922388076782, "ce": 6.176304316520691, "genuineness": 4.2368498802185055, "identity": 0.5760236933827401, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.041368373483419416, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.6570312444772022, "target_raw": 1.245346736907959, "ce": 6.166773104667664, "genuineness": 4.1904070496559145, "identity": 0.5714882865548134, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.04590378031134605, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.6277343747587127, "target_raw": 1.2707898020744324, "ce": 6.183452701568603, "genuineness": 4.206411015987396, "identity": 0.5732114553451538, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": -0.04418061152100563, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.7994140540951595, "target_raw": 1.1216934204101563, "ce": 6.168361663818359, "genuineness": 4.261953973770142, "identity": 0.5489239566028118, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.06846811026334762, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.6859374973190302, "target_raw": 1.2202429056167603, "ce": 6.170744442939759, "genuineness": 4.2836062669754025, "identity": 0.5588894145563245, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.05850265230983496, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.6609374887399266, "target_raw": 1.2419543325901032, "ce": 6.2025151491165165, "genuineness": 4.252226090431213, "identity": 0.5553120881319046, "wer": 0.008596491228070175, "valid_fraction": 1, "identity_delta": -0.06207997873425484, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.5386718737935636, "target_raw": 1.3481367349624633, "ce": 6.2219746828079225, "genuineness": 4.35734977722168, "identity": 0.6173920668661594, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 1.6960937488739927, "target_raw": 1.2114226400852204, "ce": 6.179878520965576, "genuineness": 4.318438339233398, "identity": 0.5507718324661255, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.06662023440003395, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.62265624554959, "target_raw": 1.2751999378204346, "ce": 6.185835528373718, "genuineness": 4.327852368354797, "identity": 0.5462300767190754, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.07116199014708399, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.7042968680026687, "target_raw": 1.204298585653305, "ce": 6.17114155292511, "genuineness": 4.281723463535309, "identity": 0.5784890748560428, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.03890299201011658, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.530625004160865, "target_raw": 1.355125093460083, "ce": 6.230711612701416, "genuineness": 4.354337244033814, "identity": 0.5973480761051178, "wer": 0.007453416149068323, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.7553124904128519, "target_raw": 1.1599937224388122, "ce": 6.205771636962891, "genuineness": 4.298606038093567, "identity": 0.6453817552328109, "wer": 0.009536796536796536, "valid_fraction": 1, "identity_delta": 0.04803367912769318, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.8610937478445002, "target_raw": 1.068127272129059, "ce": 6.205930452346802, "genuineness": 4.256305503845215, "identity": 0.6409243163466454, "wer": 0.009454664328806204, "valid_fraction": 1, "identity_delta": 0.04357624024152756, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.7553124904128519, "target_raw": 1.1599937224388122, "ce": 6.205771636962891, "genuineness": 4.298606038093567, "identity": 0.6453817552328109, "wer": 0.009536796536796536, "valid_fraction": 1, "identity_delta": 0.04803367912769318, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.10578125743164839, "ci95_z": [ 0.05921876212870756, 0.14937500956570038 ], "language_deltas": { "en": 0.14765625412869357, "de": 0.15468752262738542, "fr": 0.052343745871306455, "nl": 0.14218749431634392, "es": 0.032031270214512596 }, "languages_improved": 5, "clear_gain": true, "delta_ce": 0.00015881538391113281, "delta_genuineness": -0.04230053424835205, "delta_wer": -8.213220799033194e-05, "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": 0.10578125743164839, "ci95_z": [ 0.05921876212870756, 0.14937500956570038 ], "language_deltas": { "en": 0.14765625412869357, "de": 0.15468752262738542, "fr": 0.052343745871306455, "nl": 0.14218749431634392, "es": 0.032031270214512596 }, "languages_improved": 5, "clear_gain": true, "delta_ce": 0.00015881538391113281, "delta_genuineness": -0.04230053424835205, "delta_wer": -8.213220799033194e-05, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.2246874862519867, "ci95_z": [ 0.10281248955494152, 0.3835937315442037 ], "language_deltas": { "en": 0.0773437475871271, "de": 0.08359372742623397, "fr": 0.19218749774798535, "nl": 0.5320312358980988, "es": 0.23828122260048826 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.02493997573852539, "delta_genuineness": -0.05573120594024658, "delta_wer": 0.0020833803877282142, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": true, "scale": 1.0, "cfg": 5.0, "reason": "Held-out target improves against matched CFG and best manual-only configuration; QA passes" }, "reference_only": true, "updated": 1793035513.3042758, "example": { "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [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.", "language": "es", "seed": 1702, "scale": 1.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "001c9d97b1d91785525baab66f8c69d6464ae98af6af2676c4582cdaef0f8636", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f71c36d3bbe80894b7685ce4073277214a202024f47a703aed27baa9a2d67c3", "generation_seconds_batch": 25.937193818041123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6896700859069824, "target_z": 2.296875003217164, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.6483577489852905, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/001c9d97b1d91785525baab66f8c69d6464ae98af6af2676c4582cdaef0f8636.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "005c76254b5a1b397484ef8fbd8db7d6d055e1f52e876f9fdc488215cdfca02f", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e481eaa7c0bcf319bec491a9fd2c7d4279cb6de1eda3e49eeb2693800d0fa906", "generation_seconds_batch": 16.557278972002678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4224305152893066, "target_z": 1.4531249796246293, "ce": 5.903473377227783, "genuineness": 4.406930446624756, "identity": 0.6931149363517761, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/005c76254b5a1b397484ef8fbd8db7d6d055e1f52e876f9fdc488215cdfca02f.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "006a063603ce003277356fc0d883f3913b0f7d873a2a195512982b3987b0955d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c0dfedb69498b33dd08f0d3b36e3a19893ee76fb63b258d0c1f77a07992d6c7", "generation_seconds_batch": 16.463833378977142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.165581703186035, "genuineness": 4.6956281661987305, "identity": 0.5948905348777771, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/006a063603ce003277356fc0d883f3913b0f7d873a2a195512982b3987b0955d.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "0088345fd6c8b2b0bd837044fef9867af05f88746fe98ebf79cb18a486b5bd3f", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52f2f555f3dcf07be0b89e0cfafdefbeea92f29d9fe6993b3ecdf3526884410a", "generation_seconds_batch": 16.573208494926803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 6.4515180587768555, "genuineness": 4.3504462242126465, "identity": 0.4380784332752228, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0088345fd6c8b2b0bd837044fef9867af05f88746fe98ebf79cb18a486b5bd3f.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": "00aa858e5dae243ca685e47cab6692a7a6eadbedd8b1d63fcfb174f2b3a210fb", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1d0da3b402d558d68022e3abfe8ebfedb917c68e2adc5d340c7346e129850d1", "generation_seconds_batch": 22.997547816019505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.038498878479004, "genuineness": 4.545003414154053, "identity": 0.7575421333312988, "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_VOLT_reg_low/00aa858e5dae243ca685e47cab6692a7a6eadbedd8b1d63fcfb174f2b3a210fb.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "010ac0300cebf140dd8f4be943ddbd6d5387af8a08fa7fd22613663bc66bdb68", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6801ae06fa46c49c7f5c6a584ccca3a8590dee0d8c64935eae71f0c48521fa3", "generation_seconds_batch": 15.153298906981945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 5.974957466125488, "genuineness": 4.6956281661987305, "identity": 0.8004286885261536, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/010ac0300cebf140dd8f4be943ddbd6d5387af8a08fa7fd22613663bc66bdb68.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "011d915602d6b404191702b5780d58b3b90af6adde5a8e66fed9a5a8222da43f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a741425692f525e2c15ef26f0f23e013498e3d39e32e20978c061582ce71ae6", "generation_seconds_batch": 21.935332493041642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 5.935244083404541, "genuineness": 4.017816066741943, "identity": 0.5948386788368225, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/011d915602d6b404191702b5780d58b3b90af6adde5a8e66fed9a5a8222da43f.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "013cbca93026a76a5962a71b72a4ee5ff57790f275076d2428d52d91b2fcbde3", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af8caf9196e55c014737271a51c00757707564f231c4a854884124099af56642", "generation_seconds_batch": 22.26887457910925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.24500846862793, "genuineness": 4.413206577301025, "identity": 0.7824646234512329, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/013cbca93026a76a5962a71b72a4ee5ff57790f275076d2428d52d91b2fcbde3.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "0179e1dcdf990b30550ffde1b005a4daa4cab15cbb2edc58cd9c0ba0e433622d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb598aaacae0a62e7c1ad962c4ce3dc55b5c7805179636221b1bbf99d739bc8a", "generation_seconds_batch": 16.787683447939344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.449569821357727, "target_z": 1.4218749431634394, "ce": 6.125868320465088, "genuineness": 3.9927117824554443, "identity": 0.43276697397232056, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0179e1dcdf990b30550ffde1b005a4daa4cab15cbb2edc58cd9c0ba0e433622d.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "01b0e6c5e6efef5d4960f977934b4ca5e79f0ca5c7b2726f9e87ebccd1f1ceb9", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5fd2393a833181ba5856ecace6d971d9ac298e8cce489ab42a6ee2f5c4e28ec8", "generation_seconds_batch": 16.705525048077106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.427689552307129, "genuineness": 4.193545341491699, "identity": 0.6253575086593628, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/01b0e6c5e6efef5d4960f977934b4ca5e79f0ca5c7b2726f9e87ebccd1f1ceb9.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": "01dd953f26588b02e89965910b821a5e192a4091bf84a0555b43c2af07183e89", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d7f7ce770e4b3376293a68b6ad7e6e8ba0f0aef302ccb47e192f7606f1f3e76", "generation_seconds_batch": 20.588579004979692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.3641486167907715, "genuineness": 4.030367851257324, "identity": 0.399323046207428, "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_VOLT_reg_low/01dd953f26588b02e89965910b821a5e192a4091bf84a0555b43c2af07183e89.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "7b46dcbc7ab9e0e471ebc370fd079a9d4981ee504377dcca5d81ad48c8cb388d", "generation_seconds_batch": 16.08366992790252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.022613525390625, "genuineness": 4.607763767242432, "identity": 0.7038946747779846, "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_VOLT_reg_low/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "01ff069ed149febba85e804f5314efc9ac4a7f75d7327b592612778f21b35d21", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca597f448a67609b7610da46ce882b9d536b63e1d11d0208d0e0f1f5c96e388c", "generation_seconds_batch": 21.21288706699852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 5.839932441711426, "genuineness": 4.005263805389404, "identity": 0.6319193840026855, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaagt.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/01ff069ed149febba85e804f5314efc9ac4a7f75d7327b592612778f21b35d21.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "021b4dcfc9fc7377754a33bbacf2017cfae282de7760b272670512caef39f06e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83af7cfcc470b381264c8deec679ad1330bee49ede7951a281170095d28761d2", "generation_seconds_batch": 14.534111963934265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.3641486167907715, "genuineness": 4.218649387359619, "identity": 0.4470667839050293, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/021b4dcfc9fc7377754a33bbacf2017cfae282de7760b272670512caef39f06e.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "026174013a38305340e6bc0261b40e4c482a2b5dd2ed470849100b850a6b8152", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e11f093fc72a4331491d9510077ff43cfaffd1f55ea9ebae9a10ec3055503d03", "generation_seconds_batch": 15.33752458402887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 5.982900142669678, "genuineness": 4.306513786315918, "identity": 0.7823501825332642, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/026174013a38305340e6bc0261b40e4c482a2b5dd2ed470849100b850a6b8152.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "02789f93d60516a4a406ec5173dac2a3257e9b8cf3d8566065e0a87765dfe3cd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5053c478bb4668d620e62330f770b6b65c00f5578636f1719678e5e4fe946232", "generation_seconds_batch": 14.534111963934265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.348263263702393, "genuineness": 4.545003414154053, "identity": 0.47133201360702515, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/02789f93d60516a4a406ec5173dac2a3257e9b8cf3d8566065e0a87765dfe3cd.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "0335cef7706129709dd66d890f58a23ee41f288dc9b64385cdc90d0fe85a9303", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89f8e72305ca6dd2c3fe54f3749e31f99f1c200f5248b05ff4045b02586799d5", "generation_seconds_batch": 16.744034443050623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.103543996810913, "target_z": 1.820312530563056, "ce": 6.070269584655762, "genuineness": 4.174716949462891, "identity": 0.8196927309036255, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0335cef7706129709dd66d890f58a23ee41f288dc9b64385cdc90d0fe85a9303.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "0339c082937401a4f3313a40a6c86d72c6717c64898253390aa947d5087791cf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c83ab7d05054da52f349baa174f53de0824cffbdfbed47b0b8bddb8641f9cf1c", "generation_seconds_batch": 22.22923138702754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.213237762451172, "genuineness": 4.425758361816406, "identity": 0.6993838548660278, "hypothesis": "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_VOLT_reg_low/0339c082937401a4f3313a40a6c86d72c6717c64898253390aa947d5087791cf.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0357a24b4bb70439dc0219292c78955cbcb2033210b0099ba84dd31c964b0448", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caa08c7ce9a8e50c45d8cc267eeed6d03891f832893c02a442fb5253ec067582", "generation_seconds_batch": 20.834369899937883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 5.998785495758057, "genuineness": 4.20609712600708, "identity": 0.675537109375, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0357a24b4bb70439dc0219292c78955cbcb2033210b0099ba84dd31c964b0448.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "03ea25c3fd781600e512b28ecfa651520c232f7b5a085f161cc9dd96ff70cc82", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3f49b7d702f097f8f776c03be40c46f6446992c170554278d2b2b03ae26296d1", "generation_seconds_batch": 20.58476055599749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 5.95112943649292, "genuineness": 4.419482707977295, "identity": 0.6933683753013611, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/03ea25c3fd781600e512b28ecfa651520c232f7b5a085f161cc9dd96ff70cc82.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "03ea7c6c242e4ba148b72a47cd8945ccf366b6f81233b1b77a20948724bc4205", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c926f349160a3e76c7609e12653c3f720f362e1defd721babdb08e1bf7e6bd1", "generation_seconds_batch": 18.561099794926122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4631394147872925, "target_z": 1.4062499935656725, "ce": 6.062326908111572, "genuineness": 4.607763767242432, "identity": 0.774630069732666, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/03ea7c6c242e4ba148b72a47cd8945ccf366b6f81233b1b77a20948724bc4205.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "0490df3e4ae84b105d28f66285be61c993f21892a41e186ceded85ec624c77ea", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed8fd4effb24c9f34d486d5f4ca913ca174e0a1d2bf89105efc19f71a6c08ebb", "generation_seconds_batch": 15.860593963996507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.078212261199951, "genuineness": 4.21237325668335, "identity": 0.32425457239151, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0490df3e4ae84b105d28f66285be61c993f21892a41e186ceded85ec624c77ea.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "049794bcf9426b5606d760092e221b5c337a7a19eb7eda6eab6a647c0341aff1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa7a1aba7f1760963da73448305b6636ac820b9f643403e6ec9f5cb0a82bb81e", "generation_seconds_batch": 20.229721048963256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.141753673553467, "genuineness": 4.20609712600708, "identity": 0.7350454926490784, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/049794bcf9426b5606d760092e221b5c337a7a19eb7eda6eab6a647c0341aff1.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "05002e2140d232c8adf9cc6418289f28de94b0764ac98e70329b0ceb9789c198", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "644b5a233c3d4197c89c6c9a09340be037d7dc88d78f613347bf95390cd96c63", "generation_seconds_batch": 22.733477527974173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.324435234069824, "genuineness": 4.042920112609863, "identity": 0.6329141855239868, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/05002e2140d232c8adf9cc6418289f28de94b0764ac98e70329b0ceb9789c198.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "050378fa2adadb5ee8090ee4c396c2077c28657b277f607f43e2fc9d3a6e817b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7d8b6c6731f9c195d44f0f0740880bbf6b5b6bf402fa196fd02b58a7d17696b3", "generation_seconds_batch": 16.583072017063387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.109982967376709, "genuineness": 4.3881025314331055, "identity": 0.5369361042976379, "hypothesis": "Ik 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_VOLT_reg_low/050378fa2adadb5ee8090ee4c396c2077c28657b277f607f43e2fc9d3a6e817b.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "050565b06291c4da8c6d51bce033a66143c1c2ee83abf70bed2286439214c29d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f1fa92417031ef6e31b26998002cce0c72864bbe3460b73aeed91682b264692", "generation_seconds_batch": 15.68025570397731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.0146708488464355, "genuineness": 4.344170093536377, "identity": 0.6841815710067749, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/050565b06291c4da8c6d51bce033a66143c1c2ee83abf70bed2286439214c29d.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "050cc330843403c68b4e994dae9312cfce6822c99fea066e41597bd1bf7ab9ee", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b64287b120ccb6ae2ed9a86b7b089d0e121c059fab84e2866a6883a533954847", "generation_seconds_batch": 28.309883030015044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.205295085906982, "genuineness": 4.080576419830322, "identity": 0.6687212586402893, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/050cc330843403c68b4e994dae9312cfce6822c99fea066e41597bd1bf7ab9ee.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "0520f7c761d8775cf19d03df20c1df157a01ff44415dd6ea0fe8496e30f747a1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "444f8be63876a0c3cc0742a71ef1a34a4357d3c267188ff951f14c443226e303", "generation_seconds_batch": 13.748223208007403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.7003778219223022, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0520f7c761d8775cf19d03df20c1df157a01ff44415dd6ea0fe8496e30f747a1.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": "0578bbc29d0a487cbb45c84b6f2ba5c4730f37b7919744a76adc9d26b8505eb4", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7a23e1b2ce908ddd9b34592803a69fd939486258ba729d9ca35854e5f98015a5", "generation_seconds_batch": 20.83567814703565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.09409761428833, "genuineness": 4.306513786315918, "identity": 0.684569776058197, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0578bbc29d0a487cbb45c84b6f2ba5c4730f37b7919744a76adc9d26b8505eb4.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "a246dd8f531ce94a6d5e7e00f9c6725db8a5f84f739ae06c892782e6a66449e2", "generation_seconds_batch": 18.71632192307152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.374936819076538, "target_z": 1.5078124404824695, "ce": 6.38797664642334, "genuineness": 4.6705241203308105, "identity": 0.6045695543289185, "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_VOLT_reg_low/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "06250c0a081221f590b8719dea6048628c7670433d09c13a6a95c2f3a18bc5c1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e4fce67e25ec213b8ab8a57bfe3a4ccf6aa59fd7504bf5ec4e2ee31556df3bd", "generation_seconds_batch": 15.24645802390296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2392404079437256, "target_z": 1.6640624855227628, "ce": 6.125868320465088, "genuineness": 4.494794845581055, "identity": 0.6267935633659363, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/06250c0a081221f590b8719dea6048628c7670433d09c13a6a95c2f3a18bc5c1.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0651043fdf4a48ecfc7709d6c1faa274650c0ebfd7b6552d303f1325c995e4e1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2223c45fa089917c69a787a45a10876c0f8153244a1c7771512b1880f40c5770", "generation_seconds_batch": 23.336810466949828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.109982967376709, "genuineness": 4.457139015197754, "identity": 0.7697691917419434, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0651043fdf4a48ecfc7709d6c1faa274650c0ebfd7b6552d303f1325c995e4e1.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "06515711f110485a87adb9478153d2dc801bd7cfece7bfae7b25396469c54b91", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f3c0f9228ebb57c6510c2710834a3fd90287fddb068914d6f944706d4778cb3", "generation_seconds_batch": 16.00369723595213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.070269584655762, "genuineness": 4.400654315948486, "identity": 0.7765641808509827, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/06515711f110485a87adb9478153d2dc801bd7cfece7bfae7b25396469c54b91.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "06758651cb47b90cc5a0a42c27dd1fce24fb68a502f3d2e9ed8b7eb1f5629c95", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fca8e854825c8ce436d8926b6679886729bd8397d9c2bd6b9ebbe86bfe84f0a", "generation_seconds_batch": 14.900843855924904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4699242115020752, "target_z": 1.3984375187667888, "ce": 6.38797664642334, "genuineness": 4.607763767242432, "identity": 0.3787837028503418, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/06758651cb47b90cc5a0a42c27dd1fce24fb68a502f3d2e9ed8b7eb1f5629c95.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "072d9cdf9e6a392ac0f663843e24c5875fe0aeda914d958758d78106f293e552", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccd9dacfa230ade79297835399f5bd5116d795b06cffb57f245e246fcb196f91", "generation_seconds_batch": 16.358452677959576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 5.95112943649292, "genuineness": 4.162164688110352, "identity": 0.5782477259635925, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/072d9cdf9e6a392ac0f663843e24c5875fe0aeda914d958758d78106f293e552.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "074c326f4a16c1c84bb635729b429e01b0cbb9b05261dfb45fcaed6b952e2669", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22da7646ea92fd39644291025c96adc2a294ebb001c227f57bf96c90ca786299", "generation_seconds_batch": 15.603602790972218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.165581703186035, "genuineness": 4.457139015197754, "identity": 0.42545798420906067, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/074c326f4a16c1c84bb635729b429e01b0cbb9b05261dfb45fcaed6b952e2669.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "07638bd80c06ec90f1b5f99d632bf53021ae2e158519c3759ab67e8b09f67881", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32a0dc20d4e137c7b3723e07a0285efe0f3248e0b41c3cfaaf1cf0d6a56c22b6", "generation_seconds_batch": 22.5631639380008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.994986891746521, "target_z": 1.9453125391421597, "ce": 6.181467056274414, "genuineness": 4.042920112609863, "identity": 0.3002570867538452, "hypothesis": "I found the little box. It was beside the old window. We can look inside to care.", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/07638bd80c06ec90f1b5f99d632bf53021ae2e158519c3759ab67e8b09f67881.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "07f9575cd0c5bd1ed77ee91163ebaf78c835378f416eb2deed218bc47ff66172", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b3f014861c9cc9fffa42fdeb7232ec72c576f46e7386c57350864702e297f255", "generation_seconds_batch": 23.503076920984313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.372091293334961, "genuineness": 4.256305694580078, "identity": 0.6294045448303223, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/07f9575cd0c5bd1ed77ee91163ebaf78c835378f416eb2deed218bc47ff66172.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "0837b0ae99838653d83dda298131188bef2a566bb708b145deb405e393e71ab4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64c916312254c1209ddaa63fccdf839e8a1c2237922e92c56b947c08bc5e321e", "generation_seconds_batch": 21.935332493041642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.141753673553467, "genuineness": 4.15588903427124, "identity": 0.7100322842597961, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0837b0ae99838653d83dda298131188bef2a566bb708b145deb405e393e71ab4.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "085803405e0a5791cb8ea424ceb3b0eb03aa1bd8c25315d388b28ada07631743", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "781ed38627f15fba01ad2c371c3dc76ac54b263f054e3f8da1e1f280a46408ad", "generation_seconds_batch": 20.600569118047133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.0305562019348145, "genuineness": 4.030367851257324, "identity": 0.6801697015762329, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/085803405e0a5791cb8ea424ceb3b0eb03aa1bd8c25315d388b28ada07631743.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "085c8e0807d8d157ec31302d02d16b496925d3e0bbdfbda521e95079a5c4cbee", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2a270275ea39768b18bf8a926c60205a721194bb68c23cbfb24e06abf22e6f3", "generation_seconds_batch": 15.33289022394456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.38003396987915, "genuineness": 4.300237655639648, "identity": 0.2721124291419983, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/085c8e0807d8d157ec31302d02d16b496925d3e0bbdfbda521e95079a5c4cbee.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "0885fd068fde788915acdd7c48cd1ba7260134d3c0f31988540dae8477f8ef44", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46d437fbb8395976a53f251bd53f9e6c4a534cf7d0c1ce6759a9b38baa10fa76", "generation_seconds_batch": 25.222870611934923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.38797664642334, "genuineness": 4.394378662109375, "identity": 0.5250356197357178, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0885fd068fde788915acdd7c48cd1ba7260134d3c0f31988540dae8477f8ef44.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0897ec71ba84909f872cc7216e3cede0700153c518a84dd0aaf090a4826f89f5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb6dbacf915db4fa20520172e48aa496cf664535b8e36c6e884f52698ebd3641", "generation_seconds_batch": 22.966198301059194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.1894097328186035, "genuineness": 4.375550270080566, "identity": 0.5375099182128906, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0897ec71ba84909f872cc7216e3cede0700153c518a84dd0aaf090a4826f89f5.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "097630531bf4226efc3caf087a4748370e5437f9ce3f4d5372adc9520c6fb7f5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9032250bb64ae669a304980fac5807ce5ef038db5285d9b610d4c5dacf0f86c5", "generation_seconds_batch": 21.963544397964142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5377724170684814, "target_z": 1.3203124962466422, "ce": 6.24500846862793, "genuineness": 4.3881025314331055, "identity": 0.7894202470779419, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/097630531bf4226efc3caf087a4748370e5437f9ce3f4d5372adc9520c6fb7f5.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "09a4689de5efe2262936f2da7d2082d465827a3f6e8e9a4aca8b1f5efda9fb49", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6578099bb0b558a4faeebaf4407fe81cb61265e03aea3cac8f6ac4465289377a", "generation_seconds_batch": 15.340852351975627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.260893821716309, "genuineness": 4.143336772918701, "identity": 0.3632356524467468, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/09a4689de5efe2262936f2da7d2082d465827a3f6e8e9a4aca8b1f5efda9fb49.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "09b4c9a57844c058e44b96e4e51603e43488d400d885bd6da045c3567114610f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b0ce9837c5ab32b06e6380df0f4f2cd4b993c6bebb3dc00fe01abb998ad7ff5", "generation_seconds_batch": 22.204893738031387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 6.292664527893066, "genuineness": 4.243753433227539, "identity": 0.5633699893951416, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/09b4c9a57844c058e44b96e4e51603e43488d400d885bd6da045c3567114610f.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": "0a100968e72c95d714771dd2edf027554d30a9aa72525db0354bf7dddea152b1", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "76127aa3e863e6bd7ec06ae3d004e6476e318e39a6fea3675c66c99d3040cc7e", "generation_seconds_batch": 22.647255984018557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.157639026641846, "genuineness": 4.293961524963379, "identity": 0.8036097288131714, "hypothesis": "Te escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0a100968e72c95d714771dd2edf027554d30a9aa72525db0354bf7dddea152b1.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0a46132332c8efebb598324facfee3bfc61b3c6f91dbc9a5a203a9e90e1b687e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db6e22874c409300d5af2bfe30a55bac8fa5875b62f4b18b2378e8fa92f53b18", "generation_seconds_batch": 23.687269009999, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.7616239190101624, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0a46132332c8efebb598324facfee3bfc61b3c6f91dbc9a5a203a9e90e1b687e.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0a49af43509f0bfb14b362d429cdbab0b26abee6b62b52da7df9f9499011aa05", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a72ca6015341e0e4a366bc345315c7be521a4c93e583318669d8fe75069cdd7", "generation_seconds_batch": 22.966198301059194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5649117231369019, "target_z": 1.2890624597854523, "ce": 6.125868320465088, "genuineness": 4.256305694580078, "identity": 0.6013922691345215, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0a49af43509f0bfb14b362d429cdbab0b26abee6b62b52da7df9f9499011aa05.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0a5e9629792e4605f370778514fa82d6896dc89e87297a8c51aa0b6b0af575d1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdb2d7f7eeb6cd2c687bcd7cd906b75d0126ce5eb71aa921ef4cd9f5a7cc496c", "generation_seconds_batch": 23.336810466949828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 6.078212261199951, "genuineness": 4.432034492492676, "identity": 0.755776584148407, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0a5e9629792e4605f370778514fa82d6896dc89e87297a8c51aa0b6b0af575d1.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": "0a742cea65de1e30ecb4f04888c2749cd9443873212d4ebaf42a496a47f5f840", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc74ef2a5d26facb3c5701d47e18ee98faa7a099eb09d94f67b8b43b489952f7", "generation_seconds_batch": 16.07816788495984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9814172983169556, "target_z": 1.9609374887399267, "ce": 6.173524379730225, "genuineness": 4.268857479095459, "identity": 0.8325093984603882, "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_VOLT_reg_low/0a742cea65de1e30ecb4f04888c2749cd9443873212d4ebaf42a496a47f5f840.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0a86e266f29f2aa322db788aed8e04aeca70cc9edad80e51cf08012d3a24c356", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3a8fcfbebac6b7f126f62e7a5795b6e8bdbc66bc8bc980b8b0021ed5cd1087a", "generation_seconds_batch": 22.172430134029128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.181467056274414, "genuineness": 4.268857479095459, "identity": 0.7594209909439087, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0a86e266f29f2aa322db788aed8e04aeca70cc9edad80e51cf08012d3a24c356.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "0a92f93e41d72d085acd4aa5a0dba62d06b9a27cfeb9fb94b365504f2d4061cd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d1dfc9b86cc7d1c79061557d767603640899567586b2ef4e21016324355849c", "generation_seconds_batch": 20.409897082019597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.133810997009277, "genuineness": 4.582659721374512, "identity": 0.7395610809326172, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0a92f93e41d72d085acd4aa5a0dba62d06b9a27cfeb9fb94b365504f2d4061cd.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "0aeeec3fa26786c9da832415aabf9c45e01ad7de66be59dcad97613887eae260", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6a5567c420de06e91e203e46a54f4a31c50b6bc95e4ec153eee856c7d7b5872", "generation_seconds_batch": 21.15319903299678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.348263263702393, "genuineness": 4.362998008728027, "identity": 0.567418098449707, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0aeeec3fa26786c9da832415aabf9c45e01ad7de66be59dcad97613887eae260.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "0b2a1b88a5f5e92161e63f4169b201dc9edb91b15309f349289f2672a19b92b5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc24f261ff212d57b7f80353f34127f8185e0ca5cdb7cddb3b4cbfe5941c0cc2", "generation_seconds_batch": 15.149914038949646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5675432682037354, "target_z": 2.43750009865969, "ce": 6.038498878479004, "genuineness": 4.168440818786621, "identity": 0.5383192896842957, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaart.", "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_VOLT_reg_low/0b2a1b88a5f5e92161e63f4169b201dc9edb91b15309f349289f2672a19b92b5.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "0b8324e25861ecc79401c18249d8b13f2a01a165317920b5a4a65b8381c25003", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da3582b0372757364ee5c78caaa61ec31cf2fdcb427dc4a2de0d1a02b420f13a", "generation_seconds_batch": 21.210514620062895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6625308990478516, "target_z": 2.3281249024126978, "ce": 6.23706579208374, "genuineness": 3.8671910762786865, "identity": 0.3580673933029175, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0b8324e25861ecc79401c18249d8b13f2a01a165317920b5a4a65b8381c25003.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "0bbb3868763115f47b7190f666fe58132fcbd5e200e56a414e00e863fe7a5a67", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dea038c72bc631be6cd3b3f1f9f58c981ddad8d2ca832a0d7bc594122df76299", "generation_seconds_batch": 20.952234451076947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 5.855817794799805, "genuineness": 4.030367851257324, "identity": 0.6872695684432983, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0bbb3868763115f47b7190f666fe58132fcbd5e200e56a414e00e863fe7a5a67.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "0be1bf95bd7cadfa5d192380974fab98c3d2b94adc8846ca229eb392aeb29d0e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4a79271dfa100e8da9317e59dc1f1777762fae9248560276961b9da34f372bd", "generation_seconds_batch": 21.424190304009244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 5.967014789581299, "genuineness": 4.068024158477783, "identity": 0.4431186318397522, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0be1bf95bd7cadfa5d192380974fab98c3d2b94adc8846ca229eb392aeb29d0e.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "0be32258ae06e3f50e9a17d27e68ad50ffa607c65ec464d2e12ceb3f1287504a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1c39d4c8b8a31f9f59a27b069c98fbc78c8a63454fd6775b3b68da167c69f4a4", "generation_seconds_batch": 15.496189635945484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 6.173524379730225, "genuineness": 4.293961524963379, "identity": 0.9010788202285767, "hypothesis": "La porte 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_VOLT_reg_low/0be32258ae06e3f50e9a17d27e68ad50ffa607c65ec464d2e12ceb3f1287504a.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "0be3ae67e8fa5129cf668053addb9490f61e23d605bd4e468ef9da1c51172738", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5db6674bb8ec2f805e34624d0895d6d31a9237773d95e6708616b3c981a2db41", "generation_seconds_batch": 21.68426930694841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.316492557525635, "genuineness": 4.375550270080566, "identity": 0.4364987015724182, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0be3ae67e8fa5129cf668053addb9490f61e23d605bd4e468ef9da1c51172738.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "107241a05722f9a3f517d2e8d6b962e839b13138e37c1d69545320063ff81b70", "generation_seconds_batch": 15.749888869933784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.260893821716309, "genuineness": 4.482243061065674, "identity": 0.5472904443740845, "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_VOLT_reg_low/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "0cc0dd3d7d14459197f0aefe6e6d35256e0168dd1a5806aee3136ebcc22917fc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e6a750c1866bc9a5933f402d828174489ed18cba36abe4f49ddb490981dbcf9", "generation_seconds_batch": 26.89300576900132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.515059471130371, "genuineness": 4.444586753845215, "identity": 0.5768576860427856, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0cc0dd3d7d14459197f0aefe6e6d35256e0168dd1a5806aee3136ebcc22917fc.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": "0ce1531aef9e8e8173d1e7c7292c86593f4fab6ffac8a8798138a562e85295a5", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3260445518a32642d31fe52c4133a68d94601870d38f32e9eac8ef8a922d0f32", "generation_seconds_batch": 22.942117439000867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 5.998785495758057, "genuineness": 4.369274139404297, "identity": 0.7843952178955078, "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_VOLT_reg_low/0ce1531aef9e8e8173d1e7c7292c86593f4fab6ffac8a8798138a562e85295a5.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "0cf1b062dd1b84c53b38a98f318ccb0849e1ca69ed6938a43539d1e1bb2002dc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d853e078f7217f6fad36dea4b9636e8984dd8796adf552556c175d6f1eb712e", "generation_seconds_batch": 20.759771254030056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6896700859069824, "target_z": 2.296875003217164, "ce": 6.252951145172119, "genuineness": 4.005263805389404, "identity": 0.6292540431022644, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0cf1b062dd1b84c53b38a98f318ccb0849e1ca69ed6938a43539d1e1bb2002dc.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "0d06451f4f776a6180c99b747a97f03630663244026f1157c3ecf337e8c0075f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "841f3b4956f6cf5b72fb2eff64d1a47bd1757d834f7e906b1ec41f7ce69ad48f", "generation_seconds_batch": 22.45217899302952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.24500846862793, "genuineness": 4.068024158477783, "identity": 0.2675008475780487, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0d06451f4f776a6180c99b747a97f03630663244026f1157c3ecf337e8c0075f.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "0d0895e8316a33f85f97f09847bf0ad16a39ab61040a1a9ccfdb21254ace1476", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef852fb493c71782d22ea92a3a28a5995e6f302f7592533d1efcb400725038ce", "generation_seconds_batch": 25.222870611934923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.125868320465088, "genuineness": 4.306513786315918, "identity": 0.5714871883392334, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0d0895e8316a33f85f97f09847bf0ad16a39ab61040a1a9ccfdb21254ace1476.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "0db87c16ccc3b3be16d1376369394a0377d4d1eef5f0ae05ae72c5f623a383f7", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "79a5388bf47fec9fe32e84876cb00d1501f565d3cfd982fd248f18917b4da7bb", "generation_seconds_batch": 17.23903857404366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 6.252951145172119, "genuineness": 4.281409740447998, "identity": 0.6280568838119507, "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_VOLT_reg_low/0db87c16ccc3b3be16d1376369394a0377d4d1eef5f0ae05ae72c5f623a383f7.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": "0dba5cebf35b1ddca791f8d077041a7c94b74d09f6947046262df9fe6084771a", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccb025329e16cd96af9cfb45a9aaa1c9254c1e86c95b7ba0f675ca9dd2a171b6", "generation_seconds_batch": 17.022747148992494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2392404079437256, "target_z": 1.6640624855227628, "ce": 6.23706579208374, "genuineness": 4.419482707977295, "identity": 0.6644251346588135, "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_VOLT_reg_low/0dba5cebf35b1ddca791f8d077041a7c94b74d09f6947046262df9fe6084771a.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "0e3f68137d6f4926e4c3cbed4cc37d71ecfc91ea806493d398c89d4946f078ec", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3377346b4104342c60f782585985496b60e9e024d312951c900e45b9958842f7", "generation_seconds_batch": 23.519322487991303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 6.09409761428833, "genuineness": 4.268857479095459, "identity": 0.7580524682998657, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0e3f68137d6f4926e4c3cbed4cc37d71ecfc91ea806493d398c89d4946f078ec.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "0e486fc8b35e650b568cc1cea481a198a2003bacb3d078d4b7603dcbcdf8d3b5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd0cdfdb878accd35074a5547af9fe958f61134c2d866a8be2ecfce3619ee7c2", "generation_seconds_batch": 19.992995400098152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.308549880981445, "genuineness": 4.262581825256348, "identity": 0.4181342124938965, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0e486fc8b35e650b568cc1cea481a198a2003bacb3d078d4b7603dcbcdf8d3b5.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "0e5ba0cfca0a070746f4cdfb9d03c4c5d9b610abf5b91fe596b4d515013639bc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26f4b41d502179d73c8bfad10f6500add779db81a1d50b7bf9c1f0ba9423f992", "generation_seconds_batch": 22.720765540958382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.340320587158203, "genuineness": 4.281409740447998, "identity": 0.5080411434173584, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0e5ba0cfca0a070746f4cdfb9d03c4c5d9b610abf5b91fe596b4d515013639bc.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "0e5f5800ae74ecaf5d5f77c20c19f093de8d914297974a7bbacf9c9dab7dc328", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6659cbc02c3294501ab57a4a52ddfc5b7eb523edf4a4ded37d6eb2275279af7f", "generation_seconds_batch": 19.829761540982872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.229123115539551, "genuineness": 3.9425034523010254, "identity": 0.3357705771923065, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0e5f5800ae74ecaf5d5f77c20c19f093de8d914297974a7bbacf9c9dab7dc328.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "0e94d02b76cfece1e6a9f378adce9d78d73cd2b0b9c244857e3d955a9f116162", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "382e59f282bc752784672f0e63be11ce783f50c72714c182101d1f95d284008d", "generation_seconds_batch": 16.66152043407783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4224305152893066, "target_z": 1.4531249796246293, "ce": 6.109982967376709, "genuineness": 4.545003414154053, "identity": 0.6909295320510864, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0e94d02b76cfece1e6a9f378adce9d78d73cd2b0b9c244857e3d955a9f116162.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "0ec8a275a068a759f098ca931efa8ce036ea9b0e557a9e607bf8dbb167f690bc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd2ff5ebe4cb12e5f956af287610268dac2e9b4bf3cda3beac65ad2951569527", "generation_seconds_batch": 20.69444750796538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.292664527893066, "genuineness": 4.111956596374512, "identity": 0.683336079120636, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0ec8a275a068a759f098ca931efa8ce036ea9b0e557a9e607bf8dbb167f690bc.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "0ecf02645171bf6c5d95505a7ed77047d4339704d0f42b512d970a7434e3e3ba", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7666f716c99f5d78bd22c5cd130ca2ef160d836b46487e0483c8fbf0c4afa8de", "generation_seconds_batch": 20.505616639042273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 5.903473377227783, "genuineness": 3.9550557136535645, "identity": 0.698498010635376, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0ecf02645171bf6c5d95505a7ed77047d4339704d0f42b512d970a7434e3e3ba.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "0efa1ef0d7cea46885e8550b66baa4bf9dba958c2525d75a4b84c59ea90db250", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "23e92c97178d22c97715e46f125506b16d1d2f6c747b5152892824c0e168bc95", "generation_seconds_batch": 23.908192328060977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.125868320465088, "genuineness": 4.256305694580078, "identity": 0.712052047252655, "hypothesis": "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_VOLT_reg_low/0efa1ef0d7cea46885e8550b66baa4bf9dba958c2525d75a4b84c59ea90db250.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "0f8a26f6d923dd563134f7066a1902400faf49d6e25a00cf85c798681674ad79", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfb02ff8ee21eeec77c70591fddea98540c2bcdc2ad88afcb010c5c892e47808", "generation_seconds_batch": 19.829761540982872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4631394147872925, "target_z": 1.4062499935656725, "ce": 6.340320587158203, "genuineness": 4.231201171875, "identity": 0.31570273637771606, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/0f8a26f6d923dd563134f7066a1902400faf49d6e25a00cf85c798681674ad79.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "1022a1f12e0da5860136f947e260be1998fb0cd66983a17fa6ce0e2356321cfe", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cbe26fd434f958545fa16b582c998064d232b751f808a56705fad66b4fb908d", "generation_seconds_batch": 20.607824043021537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.165581703186035, "genuineness": 4.319066047668457, "identity": 0.3308265209197998, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1022a1f12e0da5860136f947e260be1998fb0cd66983a17fa6ce0e2356321cfe.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "10422b57e76bcf0d8681cf0173dff6210a505ba694b21576bd2b7949d7681130", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b66f42715212c8dba9baa3445268b9467579f35e5811c35e56244dc8af12fe43", "generation_seconds_batch": 22.17207066901028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.292664527893066, "genuineness": 3.9676074981689453, "identity": 0.5127496123313904, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/10422b57e76bcf0d8681cf0173dff6210a505ba694b21576bd2b7949d7681130.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "1060a50d81a54b72bc24f6ba5e52c795c19af405b7f5e37e07216d1feae92ca8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cd9095b2bff6974ac33e1b1208f7671e6541a3dfea5a825a845c5c893b477e80", "generation_seconds_batch": 21.3136409180006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.419747352600098, "genuineness": 4.331617832183838, "identity": 0.6405571699142456, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1060a50d81a54b72bc24f6ba5e52c795c19af405b7f5e37e07216d1feae92ca8.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "109eb0a1c7142b59e02835781eca9588758ac7ec477415d9126d4b98e53fbced", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f443077ca0e7c21049eb15c7fc7d2ad4a5e75d881d14145f1cd17c41f5e0e10", "generation_seconds_batch": 23.018887489917688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 5.879645347595215, "genuineness": 4.532451152801514, "identity": 0.7761406898498535, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/109eb0a1c7142b59e02835781eca9588758ac7ec477415d9126d4b98e53fbced.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "10bc624c7580239607cedab629cd7fa7f8e20b5007232d391a99b71b01d09eb9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fe1c22cdfb498288657857e086d7869ef89b53015c3e88f99ae25e6dc39e15b", "generation_seconds_batch": 23.962750650942326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0628350973129272, "target_z": 1.867187516622013, "ce": 6.2767791748046875, "genuineness": 4.099404335021973, "identity": 0.21092277765274048, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/10bc624c7580239607cedab629cd7fa7f8e20b5007232d391a99b71b01d09eb9.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "10d55eb50577e6ac5d77b5b83cbf884c80a6e73ee437055f83075ff6b35a8f08", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd2017e6459671d375a8eddee03fde4e8850ee7e8e12d4e270db6ceb6fae4faa", "generation_seconds_batch": 16.539025965961628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.372091293334961, "genuineness": 3.9927117824554443, "identity": 0.6427789926528931, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/10d55eb50577e6ac5d77b5b83cbf884c80a6e73ee437055f83075ff6b35a8f08.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "10d86f9bec04eef066a211f6aa78b99ac03732d1be98004ba96e12539febcd87", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7167ca7c00e3dff5903a622e5d28a3adaddb411a60df4090450131a1da549a9", "generation_seconds_batch": 20.185415750951506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 5.974957466125488, "genuineness": 4.281409740447998, "identity": 0.5768890380859375, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/10d86f9bec04eef066a211f6aa78b99ac03732d1be98004ba96e12539febcd87.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "10efb5993152a37fdc73261276d56e4fe96c91607fb554d74e4431e79af89e31", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b9e0c938c6afca685717f45db9842899af601f0c61f14e38376365b6c3c6919", "generation_seconds_batch": 23.204076971975155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.133810997009277, "genuineness": 4.18099308013916, "identity": 0.797653317451477, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/10efb5993152a37fdc73261276d56e4fe96c91607fb554d74e4431e79af89e31.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "10ff7be6acd39b6f159ad399db48d2dff2aa6e80a57496367f1276a9a742b016", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "49dc57b02c4b61cb3a6fe77bfac11c282dd821f64b44da1c9b5db2179400ef1b", "generation_seconds_batch": 21.211613881983794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9678477048873901, "target_z": 1.9765624383376936, "ce": 6.340320587158203, "genuineness": 4.369274139404297, "identity": 0.4081476330757141, "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_VOLT_reg_low/10ff7be6acd39b6f159ad399db48d2dff2aa6e80a57496367f1276a9a742b016.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "115df0b1508659a3164ff5e931a2ce33b76093e2cd3979208e26d69c4aba3833", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1cb727390f30d53a03198ab0dbe9a63b36d739a09e4a4f3cd43e7c19091f618", "generation_seconds_batch": 22.72625246993266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 5.9431867599487305, "genuineness": 4.6705241203308105, "identity": 0.6877924203872681, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/115df0b1508659a3164ff5e931a2ce33b76093e2cd3979208e26d69c4aba3833.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "117d1ccbe0b5d253c75531a1924777b81825c4b02dba4386796f8a1a2df5009d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ad5974da8344dbd7bf11b2aee5437236a0c3f9715adc6a91d6a24acc58d754a", "generation_seconds_batch": 23.668913187924773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 6.252951145172119, "genuineness": 4.1998209953308105, "identity": 0.2999172806739807, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/117d1ccbe0b5d253c75531a1924777b81825c4b02dba4386796f8a1a2df5009d.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "1198c6da827e558ed9b512f6b17ffc853e8a8c8de00c9f589e16e90f532d1c4e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f94ab822d55140b816132d8a6fd5032542887ab089b71af4bfb09c5228473a0a", "generation_seconds_batch": 23.0907962230267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.356205940246582, "genuineness": 4.507347106933594, "identity": 0.2917894721031189, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1198c6da827e558ed9b512f6b17ffc853e8a8c8de00c9f589e16e90f532d1c4e.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "123df8936efe38460cb8a3fbce8e464c4061ea6a4e7726867f45ef56deb841bb", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e66d9b9e97aa5b76c8c9e17539161419176cf1ec101511ba8a89faeb7344d8d5", "generation_seconds_batch": 15.555638399906456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.046441555023193, "genuineness": 4.030367851257324, "identity": 0.818381667137146, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/123df8936efe38460cb8a3fbce8e464c4061ea6a4e7726867f45ef56deb841bb.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "125dce8e1e02b78777e29006da2f14501380005a5d4b0b883c80b117883187bc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "c844b677326c738265b6b206d7d4a7ce45ac6abc8aaf21f1d99c04fbc057830d", "generation_seconds_batch": 21.05792363290675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.676100492477417, "target_z": 2.312499952814931, "ce": 6.09409761428833, "genuineness": 4.469690799713135, "identity": 0.7626262903213501, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle trouvait.", "wer": 0.05263157894736842, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/125dce8e1e02b78777e29006da2f14501380005a5d4b0b883c80b117883187bc.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "c6e31b8f3c6c94ca50978d45ea8a49edcb785d9ec853bb3320f2655df1d27941", "generation_seconds_batch": 15.909555285004899, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088608026504517, "target_z": 1.468750066488052, "ce": 6.149696350097656, "genuineness": 4.362998008728027, "identity": 0.821917712688446, "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_VOLT_reg_low/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "132f7f7ae0e21763838be4b9d2adf00421edf7031cf6a81038db6c5ff558591a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3603cd211ae5308977e99de6226087f50fece995960967bb2674ec2cdc56cc00", "generation_seconds_batch": 16.66152043407783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.300607204437256, "genuineness": 4.545003414154053, "identity": 0.7825100421905518, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/132f7f7ae0e21763838be4b9d2adf00421edf7031cf6a81038db6c5ff558591a.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "1347518f78e91e56430279c03489bfc8ceff7a889e676801f0429018377b1829", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00b681d029ef5d2e578a0edcf834842e31d264a03e60f56a5a56ae300004bd42", "generation_seconds_batch": 22.90327510307543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.062326908111572, "genuineness": 4.319066047668457, "identity": 0.6966174840927124, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1347518f78e91e56430279c03489bfc8ceff7a889e676801f0429018377b1829.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "13502820ed4b8363e77dfeeae52b6c99ad5ecdc82b63fec45920a5c1f13c6ebd", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54f5d3b698c4d40525f3e6ad3921536a11a853ced86fd2b0ce620cad133f7870", "generation_seconds_batch": 16.358452677959576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.046441555023193, "genuineness": 4.256305694580078, "identity": 0.6290644407272339, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/13502820ed4b8363e77dfeeae52b6c99ad5ecdc82b63fec45920a5c1f13c6ebd.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "13d78e3caf109b1e671342384c038d9e2edb47ab6d123a238caa8ec240526dbd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e2b807d815cdf4f709c120c2ffd99cc5c6a809fc6ad71cef58f30811ac1343ff", "generation_seconds_batch": 20.71053799102083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3070886135101318, "target_z": 1.5859374630026162, "ce": 6.395919322967529, "genuineness": 4.193545341491699, "identity": 0.5711148977279663, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/13d78e3caf109b1e671342384c038d9e2edb47ab6d123a238caa8ec240526dbd.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "13e060f5554a70f4033e1526a6e9a43e92fffbaa4c7135938231e21836f533fb", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c70becf6ae4da6b8248efaf8c7519e5d343c9fa898fe25b5634303d4c7519b3", "generation_seconds_batch": 16.946781497914344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.252951145172119, "genuineness": 4.413206577301025, "identity": 0.5438018441200256, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/13e060f5554a70f4033e1526a6e9a43e92fffbaa4c7135938231e21836f533fb.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "13e25ce78a511d4634cb7885e7125e4e54d1e57d42bf603077fe327d6cd0a539", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99ab23ad8fa3ed0eab1ce169f92e076483a6c3400b7400f527f248830a318447", "generation_seconds_batch": 23.148771164007485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 5.77639102935791, "genuineness": 4.030367851257324, "identity": 0.753851592540741, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/13e25ce78a511d4634cb7885e7125e4e54d1e57d42bf603077fe327d6cd0a539.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "1418aa51f6a1f92c776eed9f8656785133fc23d1ad9ab1de6a98747bbbeaabac", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ccdd04205b5f32c5fdff239ae2fb0121da6c46dc3b827f938ac3f92a1b40c95", "generation_seconds_batch": 16.691742539056577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 5.95112943649292, "genuineness": 4.344170093536377, "identity": 0.7921487092971802, "hypothesis": "The porte d'un corps à port is encore overte. J'ai posé ton manteau on the chaise. Dis-moi quand tu es prêt.", "wer": 0.5, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1418aa51f6a1f92c776eed9f8656785133fc23d1ad9ab1de6a98747bbbeaabac.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "1452db9902f2ed98d6b5296e0b3ef9d5756768e6024f8eecb39b8e5738b450f9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ebf5968377f8fca3365b79403022c825dc065d8f0d0320e66a5495fa9bd67be", "generation_seconds_batch": 18.05265281291213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 5.77639102935791, "genuineness": 4.13078498840332, "identity": 0.7355890870094299, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1452db9902f2ed98d6b5296e0b3ef9d5756768e6024f8eecb39b8e5738b450f9.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": "14849f3a945a1e8909ad42ab4be19af91b7a426d0bb5f8a29c0f85ca6a49e479", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bc0eecfb0500279ef02410a3ab07a93117eaa6c2268fb0800d2d8643fe869db1", "generation_seconds_batch": 19.348783218069002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.1894097328186035, "genuineness": 4.268857479095459, "identity": 0.6417891979217529, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/14849f3a945a1e8909ad42ab4be19af91b7a426d0bb5f8a29c0f85ca6a49e479.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "14fb56631e5651ec95d25d8c2670d069c68ee68645377fce3e31c839266303d7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfc2b66e1edba018c3627e97430bc60ca454c9331d442d285484be49bf8bb83b", "generation_seconds_batch": 15.597013700986281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.09409761428833, "genuineness": 4.005263805389404, "identity": 0.6835314035415649, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/14fb56631e5651ec95d25d8c2670d069c68ee68645377fce3e31c839266303d7.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": "151de3a908b307dd043b842053c885b2c2c30eb8ad5c12421f1d5f70525eac3e", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1084cf176cb30e57d72b66185260c40e3733ed907a2ab0499c1f6ab948cd0c45", "generation_seconds_batch": 15.71626257803291, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.221180438995361, "genuineness": 4.406930446624756, "identity": 0.5471653938293457, "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_VOLT_reg_low/151de3a908b307dd043b842053c885b2c2c30eb8ad5c12421f1d5f70525eac3e.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "2847f213f9fbdf8c44c98d0e80aa17b1bc2aed7dbd25bd2cc37bb112c67c0f4c", "generation_seconds_batch": 16.99135822802782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.268836498260498, "genuineness": 4.381826400756836, "identity": 0.6087630987167358, "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_VOLT_reg_low/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "166690994b26adf667087ae3efc9b51402c9f6c3ba9db3da0385cfbc19b5a992", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2419b0e5c412715ffb01702a76a2fd726da7adabca5d6bd9d7e01fa0c3e9be2", "generation_seconds_batch": 18.65719736693427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.197352409362793, "genuineness": 4.494794845581055, "identity": 0.6790748238563538, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/166690994b26adf667087ae3efc9b51402c9f6c3ba9db3da0385cfbc19b5a992.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "16d00e3487244ca01a43e2924224fd3c629e727799e2f03abc26fed368f30983", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0a16c1d49e9c37f67b8efeb67855ead4a5aa58423c36a8771b0c98d556bb1267", "generation_seconds_batch": 21.76490723795723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7032396793365479, "target_z": 2.281250053619397, "ce": 5.959072113037109, "genuineness": 4.093128204345703, "identity": 0.7971714735031128, "hypothesis": "La porte a encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/16d00e3487244ca01a43e2924224fd3c629e727799e2f03abc26fed368f30983.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "16f47d6633f879ef7cc1580c799f495f7e6dea775f688bf60ffa5734c4ade0b3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "424d4a72df9e861910127a872e3a4485fd293bd66d769bc9aa93e38e27517cf6", "generation_seconds_batch": 14.818628950975835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.117925643920898, "genuineness": 4.400654315948486, "identity": 0.6279492378234863, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/16f47d6633f879ef7cc1580c799f495f7e6dea775f688bf60ffa5734c4ade0b3.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "1703ee292f9cce84e64177a81808000da007954dcd8514c5f03840eb485555cd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a297b49195555a5dc98030f5cee9daca4db08721b36cccebfa314e01f6a2847d", "generation_seconds_batch": 20.834369899937883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.086154937744141, "genuineness": 4.143336772918701, "identity": 0.5915553569793701, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1703ee292f9cce84e64177a81808000da007954dcd8514c5f03840eb485555cd.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "1767ceb51d38f021e7f6c72e6c3bc64b655862531f59beecd7a4411422169b06", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f49f4fce1cdede2ed130d3f84d8459e7ad5517299484a2f5ebec174a0cfd9a27", "generation_seconds_batch": 20.952234451076947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9678477048873901, "target_z": 1.9765624383376936, "ce": 6.117925643920898, "genuineness": 4.042920112609863, "identity": 0.4932979643344879, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1767ceb51d38f021e7f6c72e6c3bc64b655862531f59beecd7a4411422169b06.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "17bd6dacbf3e9a7cc8af665314f1a2caa23931b622fead90539b77a0e355d78e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "654b016b52e95b541253b40cdb21f1a6c8564837337fb4bb35216870861bf595", "generation_seconds_batch": 23.115749009069987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2799493074417114, "target_z": 1.617187499463806, "ce": 5.998785495758057, "genuineness": 4.143336772918701, "identity": 0.46191203594207764, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/17bd6dacbf3e9a7cc8af665314f1a2caa23931b622fead90539b77a0e355d78e.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": "17d6cfed5f2430a0c2bb15ec39aa5af7e62f3a00c0e9cf9f29eb86219e136166", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a07f2fd77d5638b988f5005706909c81f23eb7ea16a47c0ec3c75cfd55fc5bd", "generation_seconds_batch": 20.83567814703565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9814172983169556, "target_z": 1.9609374887399267, "ce": 6.062326908111572, "genuineness": 4.306513786315918, "identity": 0.6246118545532227, "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_VOLT_reg_low/17d6cfed5f2430a0c2bb15ec39aa5af7e62f3a00c0e9cf9f29eb86219e136166.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "17fbadfbd4fefe33fc9e309ce1163f80bb0e7c477f15bfe28da14f3eb754671c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de8ce523fcdf64349f563fe474508f524dbfb7d9bf17535b615b4326811f42f0", "generation_seconds_batch": 20.98010233498644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1713922023773193, "target_z": 1.7421875080429094, "ce": 6.1894097328186035, "genuineness": 4.162164688110352, "identity": 0.30897682905197144, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/17fbadfbd4fefe33fc9e309ce1163f80bb0e7c477f15bfe28da14f3eb754671c.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": "18647642260d038b37f2c356f502f1973d7db2bc9128aeb5ba3edb0c169a50c5", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c41100c803d061b6ce7c0dfac7ebc0e29061e6b5f4e03c1fee05c14ff70c901", "generation_seconds_batch": 24.11141682509333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 5.935244083404541, "genuineness": 4.218649387359619, "identity": 0.7643383145332336, "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_VOLT_reg_low/18647642260d038b37f2c356f502f1973d7db2bc9128aeb5ba3edb0c169a50c5.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "186cb98911ae5fa0cd2bb3572239a7d24eb73e3bdbb7f8d3083fb8c6ad7f4655", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d52231a0414b1229334e94446eb39a0434f070f5820a4c33a1c6cc0305a1deef", "generation_seconds_batch": 21.046147524029948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 6.062326908111572, "genuineness": 4.193545341491699, "identity": 0.2821577191352844, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/186cb98911ae5fa0cd2bb3572239a7d24eb73e3bdbb7f8d3083fb8c6ad7f4655.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "1885427ccbe50524508d2cbb33a598f61ac163249d737c8a9f456d64bef60ee0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe04a92ca825fc88f9fd9106586fffb7446371aa69046e92464fa1e758ca9bd3", "generation_seconds_batch": 22.17210308101494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.022613525390625, "genuineness": 4.042920112609863, "identity": 0.5868269205093384, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1885427ccbe50524508d2cbb33a598f61ac163249d737c8a9f456d64bef60ee0.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "189f1f24c38009b2d0bcfcc3b914691759a28633518ca6a0fd2225241944d5d2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9888031fb820545a61dfa4c00257bc9603620d9b51825aaff7cfae40ae630a6", "generation_seconds_batch": 26.704046608065255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.054384231567383, "genuineness": 4.068024158477783, "identity": 0.5635286569595337, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/189f1f24c38009b2d0bcfcc3b914691759a28633518ca6a0fd2225241944d5d2.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "18cf8350f477aa362d30047990616a2043021743b5350955990e04fab9b8385f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a85553faf8180dc78654dee5b89ec3130e3e9ec769fb7a402f9c609906f8de23", "generation_seconds_batch": 23.013066195999272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.994986891746521, "target_z": 1.9453125391421597, "ce": 5.998785495758057, "genuineness": 4.18726921081543, "identity": 0.5911173224449158, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/18cf8350f477aa362d30047990616a2043021743b5350955990e04fab9b8385f.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "18f962676f7abc44fcba72da20db3327bd053300e51048e6851b093b9e428ad5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2ead4b74dbb04057c2aa3db41f4163aa1a250b38f0d8386d6e24ba1effd5779", "generation_seconds_batch": 22.777183354017325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.125868320465088, "genuineness": 4.174716949462891, "identity": 0.774556577205658, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/18f962676f7abc44fcba72da20db3327bd053300e51048e6851b093b9e428ad5.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "19062f045134c46025e29c07a7a7cf88994efd6b5ddf33453d2262e6d86e14b1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a66f3e4ba681251adda2d9a0a03f8744651e438b4706b625d5a3be7b88322503", "generation_seconds_batch": 24.015142335905693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.38797664642334, "genuineness": 3.986435890197754, "identity": 0.6285668015480042, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/19062f045134c46025e29c07a7a7cf88994efd6b5ddf33453d2262e6d86e14b1.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "19d977e2ed6466687a49597422f71ad7eb9b5e6e833e3f3c482aa1961a0a6113", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44018b4f3c01acec1509e8ea6a1eac9fe607ff5db6612fed6c229c115bd6f95f", "generation_seconds_batch": 17.756354300072417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.0146708488464355, "genuineness": 4.20609712600708, "identity": 0.4346151649951935, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/19d977e2ed6466687a49597422f71ad7eb9b5e6e833e3f3c482aa1961a0a6113.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "1a28f856398a876a7c5a381a5c72c8302d7dfaab6b4aa2b5bb02803822eeace8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b9269ffc193fe65d4da26444c05d25158b1bde4ec1cd8fafd53381664864066", "generation_seconds_batch": 15.340852351975627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.300607204437256, "genuineness": 4.093128204345703, "identity": 0.20509405434131622, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1a28f856398a876a7c5a381a5c72c8302d7dfaab6b4aa2b5bb02803822eeace8.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "1a47b309a0dff4da48d6c1926a903241bed5ec74020c24b160815889bc08bba1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dab976518ee395346927738df87e841df574d2681e977e70fd9f62661ef8f5c9", "generation_seconds_batch": 22.565550635918044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.078212261199951, "genuineness": 4.645420074462891, "identity": 0.7830229997634888, "hypothesis": "J'ai trouvé la petite boîte, elle était près de le la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1a47b309a0dff4da48d6c1926a903241bed5ec74020c24b160815889bc08bba1.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "1ad4153b9939811015b32a124a3aab050675126dcb33f2bd8a7efb3e89516c52", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a119b8f81d232fd9cfb93a512a1706d23a82bcf0322b9d1380d6472e95213d3", "generation_seconds_batch": 20.58476055599749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 5.974957466125488, "genuineness": 4.268857479095459, "identity": 0.7095423340797424, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1ad4153b9939811015b32a124a3aab050675126dcb33f2bd8a7efb3e89516c52.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "1ba1e8861ef55dcad4a33ea5893b3403151866b3354933cbd06f045145981c06", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8b12223f88bf0788eecaa8625d0e7995d9715c932351e2d006af9dd25f81c0f", "generation_seconds_batch": 14.1354167109821, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.149696350097656, "genuineness": 4.632867813110352, "identity": 0.5396026372909546, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1ba1e8861ef55dcad4a33ea5893b3403151866b3354933cbd06f045145981c06.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "1c337a63bbe9bab83e63623c6b768e106da4fc6bfc53c31b137994f686704e88", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78fb6bba61eabcb427b87b17bec481c2d7ac85a27c39c05a47d75ca8985a939a", "generation_seconds_batch": 16.41944709396921, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.141753673553467, "genuineness": 4.174716949462891, "identity": 0.8335428237915039, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1c337a63bbe9bab83e63623c6b768e106da4fc6bfc53c31b137994f686704e88.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "1c89dfd7165177c04d963782547b9aa8de9ca6ecb70f59815dd8119f5e798c9a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c128d4958c90c73dc03b411f32aacb693429c2ed4179de0cb965ba9a5dd8a96", "generation_seconds_batch": 14.632096879999153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.1894097328186035, "genuineness": 4.7458367347717285, "identity": 0.7574894428253174, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1c89dfd7165177c04d963782547b9aa8de9ca6ecb70f59815dd8119f5e798c9a.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": "1ca508c27b108a98ee165cfb7dab64134f3a219b8f8ab7bd25d969b67ebb7cad", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "787320d529115c025f70218fe91e83686c07570a0651c01ba52cdfc1d4ffa1e8", "generation_seconds_batch": 16.431593523011543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 5.839932441711426, "genuineness": 4.394378662109375, "identity": 0.8283031582832336, "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_VOLT_reg_low/1ca508c27b108a98ee165cfb7dab64134f3a219b8f8ab7bd25d969b67ebb7cad.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "1cdddbc78d29eb9f93e7004b4529a23baa8ab9ee1e21f038b4969b908cfc4024", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2176e4039ac36c4bf2b30821e15850233bcd88566a384864cd4aa8eb64e0349b", "generation_seconds_batch": 22.17210308101494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.24500846862793, "genuineness": 4.796044826507568, "identity": 0.7753176093101501, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1cdddbc78d29eb9f93e7004b4529a23baa8ab9ee1e21f038b4969b908cfc4024.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "1ce2f45d2965af7e3537e11bd414447fbc0a6b876b0bf0b5aa76782be0691575", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3a4aca57570b80934b7e92c63925bd962b3d6989683218eb657dc404436b686", "generation_seconds_batch": 23.257234496995807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.09409761428833, "genuineness": 4.137060642242432, "identity": 0.4711138606071472, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1ce2f45d2965af7e3537e11bd414447fbc0a6b876b0bf0b5aa76782be0691575.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "1d302098d7844f5558ee3b272b92679899fda291482402f2194f3f5eb869e1cd", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3224336b2a8052910510bb854a1f29012e9af92a2bec680b4d6d828d66d57d32", "generation_seconds_batch": 18.26572094100993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8634437322616577, "target_z": 0.9453124705093318, "ce": 6.4515180587768555, "genuineness": 4.6203155517578125, "identity": 0.2922372817993164, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1d302098d7844f5558ee3b272b92679899fda291482402f2194f3f5eb869e1cd.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "e6ea6cae1a39f09d889d9b21b141cc086ba99e1ec8329dc221881c98dd09cff3", "generation_seconds_batch": 16.66748444805853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3477975130081177, "target_z": 1.5390624769436594, "ce": 6.316492557525635, "genuineness": 4.099404335021973, "identity": 0.6183923482894897, "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_VOLT_reg_low/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "1d5636d43db46c3c9ce8b2bb46e0952109203611587a643d29f8f1d697a13564", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63255287c1f07c493b65e4f5fcb570e2570d4753d6378c0cd384bd70a03c91f0", "generation_seconds_batch": 20.677944466937333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.2069568634033203, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1d5636d43db46c3c9ce8b2bb46e0952109203611587a643d29f8f1d697a13564.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "75fbf3cba7d7b2c7bf99912eeb0d2faf0cfc0fc645da8547d08907c1817a930e", "generation_seconds_batch": 16.465191354975104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.109982967376709, "genuineness": 4.325342178344727, "identity": 0.6558459997177124, "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_VOLT_reg_low/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "1da1fc21e82befb8cb5fdd755630e3c514d47e9e56e6587f68e07d2eab4ba665", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "073b6a6d5978395c06a014c1189e827267dfb85e743817bcd0352757783db580", "generation_seconds_batch": 15.320706845959648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.427689552307129, "genuineness": 4.21237325668335, "identity": 0.6305071711540222, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1da1fc21e82befb8cb5fdd755630e3c514d47e9e56e6587f68e07d2eab4ba665.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": "1dbbd10fc82e0ac79c76be68ce19cfd3206cf493528dba4c456b46cb9b9971fb", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "06ef4fe10f200b162045bace196e58fe9342851c3ab1ed06e94b353b3e117a34", "generation_seconds_batch": 21.411520162946545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.622618556022644, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je wilt.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1dbbd10fc82e0ac79c76be68ce19cfd3206cf493528dba4c456b46cb9b9971fb.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "1de340026c749b202cb81a9bd54fc6dcf3a46b23374e6719937587e51670425a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1494a92acd8f3fe284e95858a6a59574fc01c1939da6aa8a4d1798622276612", "generation_seconds_batch": 25.93168057501316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5649117231369019, "target_z": 1.2890624597854523, "ce": 6.332377910614014, "genuineness": 4.287685871124268, "identity": 0.3443007469177246, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1de340026c749b202cb81a9bd54fc6dcf3a46b23374e6719937587e51670425a.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "1e75dc1e9ba237933fd9a4f61a95ab1ef544ee647131f0fbc7c82529a260823a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60213b2ebcdab95105100035d76b249422e12476e00df6bee0ff581dca0b093c", "generation_seconds_batch": 16.448964911978692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.403861999511719, "genuineness": 3.9927117824554443, "identity": 0.47800448536872864, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1e75dc1e9ba237933fd9a4f61a95ab1ef544ee647131f0fbc7c82529a260823a.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "8afcaa9a952ed0584e1b6623b525d41f7bacfbf5b699969757da69e27f060354", "generation_seconds_batch": 16.66748444805853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3342278003692627, "target_z": 1.5546875638070823, "ce": 6.24500846862793, "genuineness": 4.218649387359619, "identity": 0.7234206199645996, "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_VOLT_reg_low/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "1f22331fce74aace533964fd34dd95af2a2adf7ed3499ad8c3ae20499bee9208", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "35b472f98676e7a0ebb4f4e3930c13773cfacd98f49ccef2e2725b49b574514a", "generation_seconds_batch": 21.144257399020717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 5.974957466125488, "genuineness": 4.18099308013916, "identity": 0.6084669828414917, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1f22331fce74aace533964fd34dd95af2a2adf7ed3499ad8c3ae20499bee9208.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": "1fa4ae67436ee7a5db578e016c85856564edf45fc1d5c89fbddfcb213ca9cac6", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38481a1bb3f3074227475b59d620bb33bfbf32275839f083378646526250aaef", "generation_seconds_batch": 22.583051006076857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.38003396987915, "genuineness": 3.9173994064331055, "identity": 0.6751947402954102, "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_VOLT_reg_low/1fa4ae67436ee7a5db578e016c85856564edf45fc1d5c89fbddfcb213ca9cac6.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "1fecfe4043492cda2a418df78a17f50bbb45654b9f1e39ad7f9af7b9fa394c08", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "262556a1954d9213657e58b4270593d3cf4aa2e1d8467d931e0b8ed8e99fa54a", "generation_seconds_batch": 15.029614907922223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.340320587158203, "genuineness": 4.030367851257324, "identity": 0.30928361415863037, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/1fecfe4043492cda2a418df78a17f50bbb45654b9f1e39ad7f9af7b9fa394c08.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "20b672abdb68c688e14b0110a85297bf31e36f20002a2a606f082662355b7201", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12b9c437856bc814f8c5ff7ed90eb5cf7068e481f67eabd969c985832796e1c8", "generation_seconds_batch": 25.93168057501316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3342278003692627, "target_z": 1.5546875638070823, "ce": 6.4515180587768555, "genuineness": 4.545003414154053, "identity": 0.39635345339775085, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/20b672abdb68c688e14b0110a85297bf31e36f20002a2a606f082662355b7201.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "20c1ca87a0bb2a5a7e6fa014744b059d53d14f57a945d07b004366ff71502257", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "813445ea8440a1c734b08310299c1f6dda630d27b437c38402b29fcea1991660", "generation_seconds_batch": 19.606482703937218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.24500846862793, "genuineness": 3.8420867919921875, "identity": 0.20952142775058746, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/20c1ca87a0bb2a5a7e6fa014744b059d53d14f57a945d07b004366ff71502257.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "20d1aad1b26484dfbef066c09ef7542bf257eadea75f5683ac026392105be713", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9630196bb13f1bcedbf0b7ed7bcecc669b9294c9fa9dcd21ffa5b26181e54dd", "generation_seconds_batch": 23.154373478959315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.197352409362793, "genuineness": 4.570107460021973, "identity": 0.6003079414367676, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/20d1aad1b26484dfbef066c09ef7542bf257eadea75f5683ac026392105be713.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "20eff935f474a446dbcd84a6b92aa88c6a3ec1c73f998b9d85ca06bf051232e7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4befea8445051116d27ed3c5f38413afb3cff87fd690dea9bfa269927d9f8849", "generation_seconds_batch": 21.245176989003085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.24500846862793, "genuineness": 3.7416701316833496, "identity": 0.3317195177078247, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/20eff935f474a446dbcd84a6b92aa88c6a3ec1c73f998b9d85ca06bf051232e7.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "20f9f73dddc1d9d1b30a21d420ab7700fd9d0e37d54bef4f8db9e953d7357644", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c984a367a1b26b2538efcc1e5a473f8e6e95c48c351bcdb1e803082a800450c", "generation_seconds_batch": 20.113493154989555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.086154937744141, "genuineness": 4.13078498840332, "identity": 0.8111579418182373, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/20f9f73dddc1d9d1b30a21d420ab7700fd9d0e37d54bef4f8db9e953d7357644.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "21574c175f3b855e20c2a22981e7445bd1be653815f04db2de98f780e91f227e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54b3c36b5837b505a8cb03f946b4c6d85f6ef14be1519503e93bea41746608dd", "generation_seconds_batch": 23.294789405888878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.300607204437256, "genuineness": 4.281409740447998, "identity": 0.4565490484237671, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/21574c175f3b855e20c2a22981e7445bd1be653815f04db2de98f780e91f227e.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "2181c57491471b3cef0c8cf1afd3448640e249b6bcbeaa16a448649bd8d4ca4c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "55d98e396448b3d176bd1e1380da9d8ed0b8f5459c06d4347a8a3312ac2a008c", "generation_seconds_batch": 20.78895715798717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.648961067199707, "target_z": 2.3437501265417766, "ce": 6.070269584655762, "genuineness": 4.2751336097717285, "identity": 0.7523173689842224, "hypothesis": "La porte 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_VOLT_reg_low/2181c57491471b3cef0c8cf1afd3448640e249b6bcbeaa16a448649bd8d4ca4c.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "2206e68367612e022a3e90026b316dce93891c61baf8d28d11815b8f04f18052", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0732c0d1cb73459af1f95e00fa4c4357165d0b6b24789f0f34ac204eb4820653", "generation_seconds_batch": 16.375400315970182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 5.974957466125488, "genuineness": 4.2249250411987305, "identity": 0.7990220785140991, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2206e68367612e022a3e90026b316dce93891c61baf8d28d11815b8f04f18052.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "222a39a3c1f87b29b80b9cba6d2a85b60c2bd45585c63a4ebe57e58257d8c6e2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0744fe468991b8eae68a3e0c9c85b136bf26edd988e80765ab1aff55d31cb0de", "generation_seconds_batch": 15.803445331053808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.1894097328186035, "genuineness": 3.9550557136535645, "identity": 0.4151264429092407, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/222a39a3c1f87b29b80b9cba6d2a85b60c2bd45585c63a4ebe57e58257d8c6e2.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": "2237109725e3ec0ba88f5346e691f08c3878b8e333a85a3e51903ef8c40acdb1", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "858f899229c4de98ddf84274dc2dc897bd347eef8d81d24bbc32dab5e9f5dcdb", "generation_seconds_batch": 19.039436175953597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.395919322967529, "genuineness": 4.162164688110352, "identity": 0.35352271795272827, "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_VOLT_reg_low/2237109725e3ec0ba88f5346e691f08c3878b8e333a85a3e51903ef8c40acdb1.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "2247d9a4a75afe0d45cbb5c6cfb3fde2a054b2afd9640ef0c1daa035b5d547d5", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94eeabe57bd4d7cda797deaf3702d2cd8568e02585dca923984abd1b7a4d9785", "generation_seconds_batch": 15.94145225500688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.6456328630447388, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2247d9a4a75afe0d45cbb5c6cfb3fde2a054b2afd9640ef0c1daa035b5d547d5.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "22efbde889c0c5987306a1e4f920b85728596aa863b075cb96a3c52567e66c82", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb12976b7cfb280b236d5208cb2ed3023dcd6eb827368ae15180a5522385846c", "generation_seconds_batch": 16.04618112789467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.229123115539551, "genuineness": 4.607763767242432, "identity": 0.7213643789291382, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/22efbde889c0c5987306a1e4f920b85728596aa863b075cb96a3c52567e66c82.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "2302433402be2a215c9d82fba6fe17dd316b8cf7aaa0f5f1c6c1d7e78683201f", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a333f1298bdf6f722ed98e85eec3560264ff7daa7d2552f622b582554d741a4f", "generation_seconds_batch": 16.707084671012126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3817216157913208, "target_z": 1.499999965683586, "ce": 6.316492557525635, "genuineness": 3.9801597595214844, "identity": 0.34934699535369873, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2302433402be2a215c9d82fba6fe17dd316b8cf7aaa0f5f1c6c1d7e78683201f.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "233137b757eb5edd731c79b9da46eea3bc014c4aa4f63ce1dd0269c95eea3f99", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2bad950f4eda587d49b1b2ffe238a790ea864c49285c6fb7e77a1464dd2f175", "generation_seconds_batch": 21.656389169045724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.749138593673706, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/233137b757eb5edd731c79b9da46eea3bc014c4aa4f63ce1dd0269c95eea3f99.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "235cff91f0c9756e2be4673c600a2cc97d89ed0c835ac79e02ab5ceebc5cce3b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6468ff219ecfb623407940ce1f46cecffc4429855f460bd54c85c7a9020aa83", "generation_seconds_batch": 21.95925581397023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.648961067199707, "target_z": 2.3437501265417766, "ce": 6.213237762451172, "genuineness": 4.231201171875, "identity": 0.5049974322319031, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/235cff91f0c9756e2be4673c600a2cc97d89ed0c835ac79e02ab5ceebc5cce3b.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "24328a3fe2cfd86ba6cbd0be68cb8d077f197b602c55120a2db2b585d2e5f91d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa14420156037bf1e9aefca687b78f9f55c93671623a110230b3136b6c467fd5", "generation_seconds_batch": 16.377679045079276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4699242115020752, "target_z": 1.3984375187667888, "ce": 6.252951145172119, "genuineness": 4.607763767242432, "identity": 0.5124914646148682, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/24328a3fe2cfd86ba6cbd0be68cb8d077f197b602c55120a2db2b585d2e5f91d.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "257d29cb828bafba1d62f2d2b606cde4053c05848c96a83914cd2f82f0c01dac", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "447c33cd5e98e62f87ca0b474c53db5c13ee7540e05da29985f45c2990c6b1e6", "generation_seconds_batch": 20.024281559046358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.2767791748046875, "genuineness": 4.055472373962402, "identity": 0.7669267654418945, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/257d29cb828bafba1d62f2d2b606cde4053c05848c96a83914cd2f82f0c01dac.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "26a8dbb1eddb9c82c864fef061a016624f16a774210b001ae746c41cba5d3279", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "46f40c2b2ffbcb5fdb5586faebae0c022986a364a08dc476ccb6bd84d3d90359", "generation_seconds_batch": 21.211613881983794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.23180004954338074, "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_VOLT_reg_low/26a8dbb1eddb9c82c864fef061a016624f16a774210b001ae746c41cba5d3279.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "8087f2f90dbee99f97cdd9cc13b6e108fc635f494d3744cea17ecca3eaae9b78", "generation_seconds_batch": 17.113209042930976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.38003396987915, "genuineness": 4.432034492492676, "identity": 0.23384177684783936, "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_VOLT_reg_low/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.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": "26d586e1e4c326b0203dc97d608ee1e7fa64d6ddd2402c370534768d153c3016", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0eae5cef933e1c05ba3f691cf52ed071ad7aedb386f5dfc5081957f537732a05", "generation_seconds_batch": 22.548764117062092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.6221858263015747, "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_VOLT_reg_low/26d586e1e4c326b0203dc97d608ee1e7fa64d6ddd2402c370534768d153c3016.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "26f445878679cc945c439975950e1da99c46a7420bd6192e6a366bbedee67897", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "892b4d91eccdb725ef16fd97e4367e9954e709b8eb9fc0b193b2ba4397e208c4", "generation_seconds_batch": 22.777183354017325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.1894097328186035, "genuineness": 4.21237325668335, "identity": 0.5710945725440979, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/26f445878679cc945c439975950e1da99c46a7420bd6192e6a366bbedee67897.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "26fe72758fe8efd6242a360e10ccb0f2fa63ce32e3509c7836a8496ccad1caeb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3dfb4c73bc186b4ff10a1892a983f9f37f5c0b3c10014e2d575aaa9a303c3d19", "generation_seconds_batch": 20.627102817990817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.292664527893066, "genuineness": 4.482243061065674, "identity": 0.659624457359314, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/26fe72758fe8efd6242a360e10ccb0f2fa63ce32e3509c7836a8496ccad1caeb.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "27350708a0af97cbed7983eb64affe3c30f4f3bddfd11afd7344ff629b3a5e7b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52b62dcfde910053fbd5f132cfb49c218a3c26f9dbb9970d715bb750263e9096", "generation_seconds_batch": 22.37740361096803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 5.792276382446289, "genuineness": 4.406930446624756, "identity": 0.5152474641799927, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/27350708a0af97cbed7983eb64affe3c30f4f3bddfd11afd7344ff629b3a5e7b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "278faee7de458ab191d964647f9ee86f026619693f4d3aac27c4646fccd7f9df", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c79cce15d67f410c3cd49a31894f15418b3d079c857d1149abe8df3cbd2876c5", "generation_seconds_batch": 23.204076971975155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.173524379730225, "genuineness": 4.111956596374512, "identity": 0.8025121688842773, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/278faee7de458ab191d964647f9ee86f026619693f4d3aac27c4646fccd7f9df.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "281a6c22a37f790e8c0c83d9162ed84251f7edb356e826aad5125cc4a65f8c85", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9583e3370b1856792d8778a6d41e5e357869e623116b6a547b2467a856cd9141", "generation_seconds_batch": 21.89769274101127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.403861999511719, "genuineness": 4.017816066741943, "identity": 0.6040059924125671, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/281a6c22a37f790e8c0c83d9162ed84251f7edb356e826aad5125cc4a65f8c85.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "281ff8f379a5e555285cb8595a2ef8d1e5e38c7959421fe03b1bc4207f8a9960", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3221ba5260e8c5ea5c528b7a7ee0096eb0d729f47289c284c24d15733aee0ea8", "generation_seconds_batch": 15.94145225500688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 5.982900142669678, "genuineness": 3.591045379638672, "identity": 0.6552187204360962, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/281ff8f379a5e555285cb8595a2ef8d1e5e38c7959421fe03b1bc4207f8a9960.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "2840cb0d7acbb1074ae82576bd69b540a5b375a851d4f294a241ae63058b5b7b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e032778fb5f8967cb9985f8c16d6198abbfaeefdd10fe594b36b5c84a5a4c4e3", "generation_seconds_batch": 22.692731917952187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.24500846862793, "genuineness": 4.243753433227539, "identity": 0.7230969071388245, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2840cb0d7acbb1074ae82576bd69b540a5b375a851d4f294a241ae63058b5b7b.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "28857ff0e8d5b7e5be7e75857d56dbc21ae08ac8165fc39698056a1ae1c68414", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7528ae155dfafdb0c1ac57e1d29e74bd48195e2ad35a33e1c4576156556c9d77", "generation_seconds_batch": 21.545071269967593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.125868320465088, "genuineness": 4.231201171875, "identity": 0.8117038011550903, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/28857ff0e8d5b7e5be7e75857d56dbc21ae08ac8165fc39698056a1ae1c68414.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "28c54901b3afe5ce447e658026a3131be3980ff5c4ac66ec5c7168d1ff265014", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf9777868b1ebf288fbcb2cdc4f497d92f4a94dddedc367887e0f058d2d35d51", "generation_seconds_batch": 15.25579786801245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.268836498260498, "genuineness": 4.344170093536377, "identity": 0.5613586902618408, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/28c54901b3afe5ce447e658026a3131be3980ff5c4ac66ec5c7168d1ff265014.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "293243a8b9ce9f45e196040ef3296d9ed6318e3480f4a8e58a18cea8f7853ec1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2203798290c454fdaa3f1d295fb9b52c96f8b5c744ceaf42c287b656ad620fc5", "generation_seconds_batch": 21.0266902979929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3477975130081177, "target_z": 1.5390624769436594, "ce": 6.125868320465088, "genuineness": 4.256305694580078, "identity": 0.704755425453186, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/293243a8b9ce9f45e196040ef3296d9ed6318e3480f4a8e58a18cea8f7853ec1.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "296ec3db067a222a0a1fcc24e856d617db41814f78558da8cd653e55b0cbc0b8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b2cdc46d8f590e90e7b8c49f45df61b1cd5539f578295b0b4e0690b493e9fd3", "generation_seconds_batch": 21.437006674939767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 6.372091293334961, "genuineness": 4.394378662109375, "identity": 0.3652941882610321, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/296ec3db067a222a0a1fcc24e856d617db41814f78558da8cd653e55b0cbc0b8.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "29a243f1f59d418582b41360897b91da1d5dbf9eaf67fbe32828aeb44d637f32", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10b45134ed859b219c22418f312c99b26a51370d6c91a15e95ff04bdf3c5b3ba", "generation_seconds_batch": 15.291850434965454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.324435234069824, "genuineness": 4.519899368286133, "identity": 0.6706055402755737, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/29a243f1f59d418582b41360897b91da1d5dbf9eaf67fbe32828aeb44d637f32.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": "29bec3ae16bfd3c93a38b166264fdb2f776da28e62575f9efb8aafbd1a23d660", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "537276e1bb8f2ebe8f8d4f7fcbf779336611dd93dfbe436174d2528b4d4b3755", "generation_seconds_batch": 22.942117439000867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.070269584655762, "genuineness": 4.093128204345703, "identity": 0.7534520626068115, "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_VOLT_reg_low/29bec3ae16bfd3c93a38b166264fdb2f776da28e62575f9efb8aafbd1a23d660.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": "2ac4242fd97f914ff925f6e4bc35d3005b351dc5362f0eeb9834d8310d117683", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d2c4b0d3761e9dff7cf747302b9861ca71aaa96f1b798ab93d65460c03a052c4", "generation_seconds_batch": 15.074642130057327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 6.0146708488464355, "genuineness": 4.011539936065674, "identity": 0.6873874664306641, "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_VOLT_reg_low/2ac4242fd97f914ff925f6e4bc35d3005b351dc5362f0eeb9834d8310d117683.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "2afa418b963acb47a62ad4014fd218e6a5a646130e5798215623b550d67aa2c2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "baaadba4be4f407085da36aac8a3a30c49884eb323fab07530fb9e1c203bfaf1", "generation_seconds_batch": 24.4185386080062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.316492557525635, "genuineness": 4.193545341491699, "identity": 0.5319977402687073, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2afa418b963acb47a62ad4014fd218e6a5a646130e5798215623b550d67aa2c2.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "2bc00c80dfc2ffcd9693174028bc1505e222cc2e17dc6956aeb6c24d75f277f6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53deb5afd2e98711a746e1b2218cd67b68deb39b63246b42d2325adc1bb8f613", "generation_seconds_batch": 23.485427560983226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3885064125061035, "target_z": 1.4921874908847026, "ce": 6.252951145172119, "genuineness": 4.65797233581543, "identity": 0.5877872109413147, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2bc00c80dfc2ffcd9693174028bc1505e222cc2e17dc6956aeb6c24d75f277f6.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "2bff262a044d6c56592f14e2dce51005a476daa4957e00ecaca366dd21d7534e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1fce3e05872d908302dc0228238ca9831cd53f4d552c1f451cb05219bb3ca29", "generation_seconds_batch": 20.65716450405307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 5.71682071685791, "genuineness": 3.8922953605651855, "identity": 0.7556813955307007, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2bff262a044d6c56592f14e2dce51005a476daa4957e00ecaca366dd21d7534e.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "2bffdf13df4f78b5fe0ac7b1f69dbe367dd6f2aad99b2a08f288a539253503ce", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caeb7da30b8518724dece3fdc1e63ed104dcb25de65f20ad6b956597f07faf3d", "generation_seconds_batch": 22.67651059210766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.038498878479004, "genuineness": 4.18099308013916, "identity": 0.6060545444488525, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2bffdf13df4f78b5fe0ac7b1f69dbe367dd6f2aad99b2a08f288a539253503ce.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": "2c2bbd678f5c321ed6f491b94824602f66bdc9d09326cfdea4b7bee6e7ce670a", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45a1d422ac0b59da65d5cbb5eb1c8f8e9035f527c7acecdfcb2dcf3c5cdd4b7f", "generation_seconds_batch": 14.38022274908144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.356205940246582, "genuineness": 4.017816066741943, "identity": 0.30629801750183105, "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_VOLT_reg_low/2c2bbd678f5c321ed6f491b94824602f66bdc9d09326cfdea4b7bee6e7ce670a.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "2c303da2835236d8cb9efba925343df9b5fcb04664abea39a01f6549fce63d2d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cedfc8ec3f0ca4dddf557a8c5a78e4c326c34161e681934f35802ce0ead2bf2f", "generation_seconds_batch": 20.946646210039034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.316492557525635, "genuineness": 4.319066047668457, "identity": 0.7369799017906189, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2c303da2835236d8cb9efba925343df9b5fcb04664abea39a01f6549fce63d2d.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "2c954ce949efcaa457d79f8d314727158f81c80ca55ca09eed1e181041a81aaf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4280bce6996c26cfd0ffae5e6fb04e79beada2610a5183d43f819c2ebafc496a", "generation_seconds_batch": 17.247034652042203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.141753673553467, "genuineness": 4.149612903594971, "identity": 0.7620100975036621, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2c954ce949efcaa457d79f8d314727158f81c80ca55ca09eed1e181041a81aaf.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "2c99e8fbada901941b4f5e193092e9de879734dc7ccda9865e3a152823bbf8b9", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65a7fb75f13a7037fd6af361d76776fb76db2c3a8aa8a68ea08e42deb4a016d8", "generation_seconds_batch": 18.03892247704789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5784813165664673, "target_z": 1.2734375101876854, "ce": 6.141753673553467, "genuineness": 4.6956281661987305, "identity": 0.6636307239532471, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2c99e8fbada901941b4f5e193092e9de879734dc7ccda9865e3a152823bbf8b9.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "2c9c5b7f06b1b34d47eaa3123eb3cf33de4ad015b6ca073941a50f0f8ac232d8", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49a144c30f04aaa5b3c05395ace210d87e25db605de40cc8cbf49711c19fabc3", "generation_seconds_batch": 15.74778386799153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7752411365509033, "target_z": 1.046874917426129, "ce": 6.221180438995361, "genuineness": 4.2374773025512695, "identity": 0.7140368223190308, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2c9c5b7f06b1b34d47eaa3123eb3cf33de4ad015b6ca073941a50f0f8ac232d8.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "2d5634686c76fe4a20308f8e5ee147896f4c00d6878db58f796b7d9a55f23acc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "895383b1e832af53f5fdedc3f391777c5716cfde64c83218f46b49a69e2bc5fc", "generation_seconds_batch": 23.962750650942326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.316492557525635, "genuineness": 4.413206577301025, "identity": 0.33446747064590454, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2d5634686c76fe4a20308f8e5ee147896f4c00d6878db58f796b7d9a55f23acc.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "2e4fe8479b9877ec2ca419b734571141b7e24f20b19309db79e2bbe81d657a8d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "565dc70620fa900b10e378ecc0686e55ea0e9899bad79c020692501aa22fd84f", "generation_seconds_batch": 22.37527203001082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5445572137832642, "target_z": 1.3125000214477587, "ce": 6.157639026641846, "genuineness": 4.65797233581543, "identity": 0.7367976307868958, "hypothesis": "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_VOLT_reg_low/2e4fe8479b9877ec2ca419b734571141b7e24f20b19309db79e2bbe81d657a8d.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "2e799bf0e1dbfb1b0effeba53d558c09eab2fa048ba0f70dfd7c83454b34c7bb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef6e20f913f4c1b8e2ee6aa06631c7c6e5ce2db4e36901d3682522207cfea9a4", "generation_seconds_batch": 21.9926674090093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.103543996810913, "target_z": 1.820312530563056, "ce": 6.260893821716309, "genuineness": 3.923675537109375, "identity": 0.47790592908859253, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2e799bf0e1dbfb1b0effeba53d558c09eab2fa048ba0f70dfd7c83454b34c7bb.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "2e8d956c89c4deaabdf74a708eebafb1f041c905781d14e0bfe69848a0abd062", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "858e4fb80f70a8f4f9b876754687545549526319226c69c0ed45a08d3e28b173", "generation_seconds_batch": 16.818142972071655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.38003396987915, "genuineness": 4.432034492492676, "identity": 0.31293222308158875, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2e8d956c89c4deaabdf74a708eebafb1f041c905781d14e0bfe69848a0abd062.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "2f3b90c13f1cd39ae456d8b1a56470f6a695ce30306dd83f15b7fdbb9f3638a3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac3cbc6e5e1b4a118f9731b3214192ab9607c31954defa25e89c852eeba7cefa", "generation_seconds_batch": 21.074272376019508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.324435234069824, "genuineness": 4.231201171875, "identity": 0.05193120613694191, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2f3b90c13f1cd39ae456d8b1a56470f6a695ce30306dd83f15b7fdbb9f3638a3.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "2fb7b8008725a219e57196c72b2146f6212bc91147d08863b5133b565e29681b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92ddd5a83a108eca411f5f95dff655dc59756be56521ebbe11d3b48ccba42be5", "generation_seconds_batch": 24.889232890098356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0356957912445068, "target_z": 1.8984375530832027, "ce": 6.022613525390625, "genuineness": 4.582659721374512, "identity": 0.8140665292739868, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/2fb7b8008725a219e57196c72b2146f6212bc91147d08863b5133b565e29681b.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "30c7e3a0bea095890fdc6bd3680acad6b1ed626e4f71ff603e9e706248d8d717", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a24dccedc37298348aeb65011805d3f9b2f853059229a0173dc502069c1c4160", "generation_seconds_batch": 16.06397369899787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.229123115539551, "genuineness": 4.281409740447998, "identity": 0.4346253573894501, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/30c7e3a0bea095890fdc6bd3680acad6b1ed626e4f71ff603e9e706248d8d717.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": "314473a397d8c77c6cde3dcada43cbedcca62e052af529615bf4b19bd41f4934", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e80b013181a3f62fb0f403e6a36364f146655591af1fde0dc643d4929c35666", "generation_seconds_batch": 17.552321463939734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.538887023925781, "genuineness": 4.319066047668457, "identity": 0.5163773894309998, "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_VOLT_reg_low/314473a397d8c77c6cde3dcada43cbedcca62e052af529615bf4b19bd41f4934.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "31734500abb778ba853555a5172e296f82665119cc83fe21e38224fb66b3d448", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63c21ddded64ebc09c4d74939ebff54869109c4b46d8493d15138adb8ef37516", "generation_seconds_batch": 17.26126123801805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 5.77639102935791, "genuineness": 3.7165660858154297, "identity": 0.6326799988746643, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/31734500abb778ba853555a5172e296f82665119cc83fe21e38224fb66b3d448.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "3176c02711277d353fedaac83911c568f556a8055ae906e0113c32276415937c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28d046557c2225b1b84481c5ef3cd6b7bd9153ca50302782012d6a813e6c8585", "generation_seconds_batch": 22.50529301003553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 5.974957466125488, "genuineness": 4.055472373962402, "identity": 0.6339284181594849, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3176c02711277d353fedaac83911c568f556a8055ae906e0113c32276415937c.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": "317baa1dea53edb7172c54cbed1707f0235670d6ce86231655ca337c40106a47", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "219b884ae46be61f2b635c140b88e55e159a2ea07504301587a45fcfb3abbf79", "generation_seconds_batch": 17.054749927017838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.133810997009277, "genuineness": 4.444586753845215, "identity": 0.6194036602973938, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/317baa1dea53edb7172c54cbed1707f0235670d6ce86231655ca337c40106a47.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "3193ab093cf1b81f0954dd9fabc00d88273b70d869aad8155dccabb03d4d0205", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f34666438d5cf4079aa3d50a035c06d938b7e995837e1ac6502f88a63537b954", "generation_seconds_batch": 22.97760985395871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.09409761428833, "genuineness": 4.519899368286133, "identity": 0.7972941398620605, "hypothesis": "J'ai trouvé la petite la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3193ab093cf1b81f0954dd9fabc00d88273b70d869aad8155dccabb03d4d0205.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "31f9a60ad37c2a12726264cb5e8149c18cf8734ed4abbd89ff588eedddce777f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19219b0fcbeb45c311183fad56f320924bfb33fe4dd0c04be8b1d4cc7c710039", "generation_seconds_batch": 20.561676584067754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.062326908111572, "genuineness": 4.432034492492676, "identity": 0.7162474393844604, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd: zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/31f9a60ad37c2a12726264cb5e8149c18cf8734ed4abbd89ff588eedddce777f.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "323ad66d44f4e93fbd68daea27ad7d4979939fde04f0842738dd80a7c51042c4", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82c210139ef21c02374fb89184ec6cacba7fe1ac2bc890d92afe04bebba57794", "generation_seconds_batch": 16.833947304985486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3206582069396973, "target_z": 1.5703125134048492, "ce": 6.316492557525635, "genuineness": 4.607763767242432, "identity": 0.31062597036361694, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/323ad66d44f4e93fbd68daea27ad7d4979939fde04f0842738dd80a7c51042c4.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "329747a9cda17576d5d343fab3887b63a8fef21b4b20e3170e0603420b17a2f0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d2bd37ff228c26777dcf15582ebd17c60a7995c3caf8cb462ee3289020d9289", "generation_seconds_batch": 23.417324324022047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.149696350097656, "genuineness": 4.250029563903809, "identity": 0.6679576635360718, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/329747a9cda17576d5d343fab3887b63a8fef21b4b20e3170e0603420b17a2f0.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": "33078f6278fc7f9eea574027b2f45a5f07e06059176577e01448eb02f6ae41e1", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4d5f65c82a16dfa477faaeef8c50feb8b8ed5dfc733e7ac7aefdaa3d59306cf", "generation_seconds_batch": 15.835111622000113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.149696350097656, "genuineness": 4.243753433227539, "identity": 0.8297439813613892, "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_VOLT_reg_low/33078f6278fc7f9eea574027b2f45a5f07e06059176577e01448eb02f6ae41e1.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "334545d96c6a585870b281cfbe669532bfc511c642d26c8f1c3576947ae51af5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e64e3a6ee0f52369d80c482460d187df218ca14bad6e802cb6c255adee65088c", "generation_seconds_batch": 24.273272639955394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.117925643920898, "genuineness": 4.306513786315918, "identity": 0.7008838057518005, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/334545d96c6a585870b281cfbe669532bfc511c642d26c8f1c3576947ae51af5.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "3373754c723268dc183fc0215f98315884a1d4589678d10373088b2ce705d15e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7a3095ce8d5b84729a209afefcbba4d1cdaaa6ca1d03014d730c647a484c345", "generation_seconds_batch": 21.022987902979366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.6957063674926758, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3373754c723268dc183fc0215f98315884a1d4589678d10373088b2ce705d15e.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "33a21020d0abcee1974853491b5bbcdcfc77793e65f02429b18651a0a4ff15d8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a66b8532bdea4639757e71551a3348738499a1d9ed04912efaeca3ec1d7e90a0", "generation_seconds_batch": 20.271699463948607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 5.967014789581299, "genuineness": 4.419482707977295, "identity": 0.7740932703018188, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/33a21020d0abcee1974853491b5bbcdcfc77793e65f02429b18651a0a4ff15d8.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": "33aca44d832266c05db5caf317062e6d735de11cad55b9ec83105dfba223295c", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96dd74cff218558b91e55059bbc40ceba17e2fc302dc86cf4b12cb9d33449401", "generation_seconds_batch": 22.71756419702433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.070269584655762, "genuineness": 4.344170093536377, "identity": 0.7339295148849487, "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_VOLT_reg_low/33aca44d832266c05db5caf317062e6d735de11cad55b9ec83105dfba223295c.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "9be4ddaa2e14bbda8bd26270c9346534b3853fd56de4719ab7bbe9cce0727c54", "generation_seconds_batch": 18.00348633399699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.402076005935669, "target_z": 1.4765625412869354, "ce": 6.054384231567383, "genuineness": 4.356721878051758, "identity": 0.578191876411438, "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_VOLT_reg_low/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "3450abfb2a4d6aa785d2287bf7295a4ec5988d8fe42d80eb3f5e381058f6b78a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c61851874b991045067fad31e7b04a230599f415935e8973b9c894991a8f99f4", "generation_seconds_batch": 13.748223208007403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.173524379730225, "genuineness": 4.406930446624756, "identity": 0.728359580039978, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3450abfb2a4d6aa785d2287bf7295a4ec5988d8fe42d80eb3f5e381058f6b78a.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "2c64768fbf10e84822f0d42e318f71b69455c9df0b1a8761717f8333bf1a6528", "generation_seconds_batch": 16.957193318055943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.308549880981445, "genuineness": 3.8546390533447266, "identity": 0.2896830141544342, "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_VOLT_reg_low/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.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": "348bce606c41aac89cdcf820462f40ffc028b6c0fd00aa96c2eb592f7ea76c60", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33f1ef484ede0f43e689b56db00971430649e96f9bd3b184b7e0fed6cdf4d932", "generation_seconds_batch": 14.814975772984326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.229123115539551, "genuineness": 4.13078498840332, "identity": 0.536703884601593, "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_VOLT_reg_low/348bce606c41aac89cdcf820462f40ffc028b6c0fd00aa96c2eb592f7ea76c60.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "348cccbd4b43a14f7a319a8e5be9945cc6992e4a7922a88014fee2362f55b12e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44bb95a500882cb7959a90fc736d59a511364c6ed7ef0f04ace4ec6f5cb95111", "generation_seconds_batch": 22.77604750590399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.0305562019348145, "genuineness": 4.457139015197754, "identity": 0.695724368095398, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/348cccbd4b43a14f7a319a8e5be9945cc6992e4a7922a88014fee2362f55b12e.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "34cb22abce26cffd4b953375c3cf72ef1c0c30820792eba870ee337daf573655", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25af0fc4d4cee382f4ad2c9bb5444a28ae278b774a912ff219e3c647cb50fe16", "generation_seconds_batch": 20.65716450405307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 5.808161735534668, "genuineness": 3.8922953605651855, "identity": 0.6827127933502197, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/34cb22abce26cffd4b953375c3cf72ef1c0c30820792eba870ee337daf573655.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "3527a00cdb5053f09a5a14f6419982c1707568ec01c3786cb867685e2341c0ee", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "27fe17b26b6bda1edf38c50603e9786e514e35a75051e0fd64345be3bfa34662", "generation_seconds_batch": 16.75457545206882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 5.8875885009765625, "genuineness": 4.457139015197754, "identity": 0.6669384241104126, "hypothesis": "He encontrado la cajita. Esta junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3527a00cdb5053f09a5a14f6419982c1707568ec01c3786cb867685e2341c0ee.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "352a72f5a63145be6a752b3d01dc4928028d3174a8e7d077dc5e8619a16fe7c3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "26025d26472a76c5b36d60b723dcb5999281fe9125afda50be0898ea24fd7f48", "generation_seconds_batch": 16.676845018053427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.38003396987915, "genuineness": 4.394378662109375, "identity": 0.49211299419403076, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/352a72f5a63145be6a752b3d01dc4928028d3174a8e7d077dc5e8619a16fe7c3.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "35314b72bc400ae86ee3f1d69997d3bc88ab74a40ff6bf01581295ea103a002e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3e21cb05b5f1f4bef3b910a69f198c91b88b90aa2628739eeb2bd9861138a2a4", "generation_seconds_batch": 15.83743646601215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.054384231567383, "genuineness": 4.2249250411987305, "identity": 0.62067711353302, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/35314b72bc400ae86ee3f1d69997d3bc88ab74a40ff6bf01581295ea103a002e.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "3560a5e6e269030375c2c8687ed99b798233bf956fb23ee873503fc6a2763b7a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "423e3e4e5c771fb138c8911cd21805f6ca5e193804190cb3f00d9e7441657a28", "generation_seconds_batch": 15.527029410004616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.292664527893066, "genuineness": 4.6203155517578125, "identity": 0.6389715075492859, "hypothesis": "La porte 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_VOLT_reg_low/3560a5e6e269030375c2c8687ed99b798233bf956fb23ee873503fc6a2763b7a.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "356950082d16a916019ea07f2607cdf491953f17d7211cc533e4c12e43e1ec55", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ce92130e9e17174affbf1587adc166e6c596f608dff991503ae8983186636e5", "generation_seconds_batch": 17.010945720016025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 5.998785495758057, "genuineness": 4.7834930419921875, "identity": 0.8740915656089783, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/356950082d16a916019ea07f2607cdf491953f17d7211cc533e4c12e43e1ec55.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "35724f80b8286b395c9a1a16ce3d17ba83fa39c7f65648e8dff5d6c37efd5da9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71f5a19ac23a1469d72e7d654df24d908cb7ef383d513137e769f5437d7590f3", "generation_seconds_batch": 14.62107526208274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 5.879645347595215, "genuineness": 4.68307638168335, "identity": 0.6674855947494507, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/35724f80b8286b395c9a1a16ce3d17ba83fa39c7f65648e8dff5d6c37efd5da9.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "35a0ca839a8c7e19d7b6e6871e68d5fd82cf05d08f2a6c4676976728bce74f09", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d521069b1b26175789182c479d3e745c7309800c3eb7e199d1f4d32a43b8736e", "generation_seconds_batch": 21.63971275696531, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 5.657250881195068, "genuineness": 4.030367851257324, "identity": 0.7598081231117249, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/35a0ca839a8c7e19d7b6e6871e68d5fd82cf05d08f2a6c4676976728bce74f09.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "35cb140c00bfd3a6ac3b4f23343142b5a571f5de26f13dd4b136fd9c7717da6f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbd9ed205113bf173337cda539b1fdfd6c87f0928f91d60d59eb2295613eb9fd", "generation_seconds_batch": 21.251775570912287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.181467056274414, "genuineness": 4.042920112609863, "identity": 0.6748021841049194, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/35cb140c00bfd3a6ac3b4f23343142b5a571f5de26f13dd4b136fd9c7717da6f.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "361052c927e002a7bc41c98b4d9d09916e9a55683641ef1170616620cabd7946", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7767fa02798f1ddfc1e364c745d4bb1303c7410e680cd95bd0b75e16fa8b5a20", "generation_seconds_batch": 15.834455555072054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.213237762451172, "genuineness": 4.632867813110352, "identity": 0.7498208284378052, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/361052c927e002a7bc41c98b4d9d09916e9a55683641ef1170616620cabd7946.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "366dadb748d506d58de397aa7b1ad829a5ab6df04856d830eb975b7afb0aaff1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "75ae690c7716f97ff369ca4f87522d224a13c7a64cee0117046484880031a583", "generation_seconds_batch": 21.071168583934195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 6.0146708488464355, "genuineness": 4.519899368286133, "identity": 0.8819686770439148, "hypothesis": "La porte 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_VOLT_reg_low/366dadb748d506d58de397aa7b1ad829a5ab6df04856d830eb975b7afb0aaff1.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "3678e3d21d8502e8cdd5b952118140561c445dd4372cc34892b516b2beac6397", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "002e133a6263680649d996cb925d601f893494470159f16da51796ba3408beda", "generation_seconds_batch": 22.1537480809493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.260893821716309, "genuineness": 4.369274139404297, "identity": 0.7807216048240662, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3678e3d21d8502e8cdd5b952118140561c445dd4372cc34892b516b2beac6397.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "36ac928764aabd6b61b54979ea2cddf96198658062ff0fee0a088c45b3a8cd3a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40a418c0bb00683b0483207088902fb063ce513f952801798fe8869cc486d9c3", "generation_seconds_batch": 19.87145583704114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.141753673553467, "genuineness": 4.6956281661987305, "identity": 0.6369329690933228, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/36ac928764aabd6b61b54979ea2cddf96198658062ff0fee0a088c45b3a8cd3a.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "37386521e66e44752a03d6f286be56ab8613b88692fa1d42c3062d49659e80b4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10fb771df8b8e838a5735db6cfcf00a88a75f62a208ba02f4a1edd03abb3a8f1", "generation_seconds_batch": 22.37527203001082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.141753673553467, "genuineness": 4.306513786315918, "identity": 0.6362380981445312, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/37386521e66e44752a03d6f286be56ab8613b88692fa1d42c3062d49659e80b4.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "37a0e8498bb30e63558a62db374c1e1aafdc9fef72e73fc2011cdf1ef12531b3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d63fc682778452e8c3ae0b8c96f1e0b0051dd79aaafd590786dff1da7c275185", "generation_seconds_batch": 24.254362634033896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7710878849029541, "target_z": 2.2031250310992503, "ce": 5.998785495758057, "genuineness": 4.3127899169921875, "identity": 0.7571593523025513, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/37a0e8498bb30e63558a62db374c1e1aafdc9fef72e73fc2011cdf1ef12531b3.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": "3852d05e7618fa390ab8ec1b13b82ed045a04c25b2b1fdfdc25575807f1970a3", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af806968a75d1d3b4c9a255c8aa8f5332717ee9ef3608bcb2e8836715fc5cb5a", "generation_seconds_batch": 15.254258602042682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.395919322967529, "genuineness": 4.111956596374512, "identity": 0.23745623230934143, "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_VOLT_reg_low/3852d05e7618fa390ab8ec1b13b82ed045a04c25b2b1fdfdc25575807f1970a3.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "3880a6fdf1f16fa1033f45d3b9fc0db81aa3974e2507a994fd90b79e635acff6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd99b60392f1611cc951c82f25bf3dd9fea1abf57713602be200c41d69059801", "generation_seconds_batch": 21.341055557015352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.197352409362793, "genuineness": 4.262581825256348, "identity": 0.45031237602233887, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3880a6fdf1f16fa1033f45d3b9fc0db81aa3974e2507a994fd90b79e635acff6.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "38b900ff1904471296c5818243c2dcbd32bbb068c12209bbf10d720c2d8650fa", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3cad2dff71ef899a4cca8735a734e9bc7475042009242f1c9eb6f103c49baa6f", "generation_seconds_batch": 17.235289198928513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.316492557525635, "genuineness": 4.369274139404297, "identity": 0.4394766688346863, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/38b900ff1904471296c5818243c2dcbd32bbb068c12209bbf10d720c2d8650fa.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "38cbb9fa8a8cde15a17acef29c82a59514422a63f5257cfce97374ae3efbc75b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "71b6ac4034d574a783dc176c88d281ffc1053f282e5629879986e4b0240e87c2", "generation_seconds_batch": 21.94206143799238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.078212261199951, "genuineness": 4.582659721374512, "identity": 0.6549756526947021, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que ha dicho.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/38cbb9fa8a8cde15a17acef29c82a59514422a63f5257cfce97374ae3efbc75b.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "38f807dce2b32568bd64b7733c8fd3b54e212828568eb336c02c890c79520a83", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d360964bc61446471704db67dfe51623e05cd343e54c463f605732a01a33a7e7", "generation_seconds_batch": 15.978407702990808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.141753673553467, "genuineness": 4.105680465698242, "identity": 0.3709796071052551, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/38f807dce2b32568bd64b7733c8fd3b54e212828568eb336c02c890c79520a83.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "390e0e98f46bc779af92fb67634c82da686ebcc1792f671a9afd21533aa8d6b9", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a863c5134120001c1eea1ef7686e105e66c0ee45573328689b23119731b6acbe", "generation_seconds_batch": 15.735956192016602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.038498878479004, "genuineness": 4.143336772918701, "identity": 0.6238527297973633, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/390e0e98f46bc779af92fb67634c82da686ebcc1792f671a9afd21533aa8d6b9.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "39121179aa791ca4095dc83bd56d4b534129d8c4abf92cba9682edae8aa24fd5", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c84afba879d1ae5ee9854e41a454fb59660c0e6ce2197b626bb1a1a2f88dc34d", "generation_seconds_batch": 18.561099794926122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2935189008712769, "target_z": 1.601562549866039, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.50908362865448, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/39121179aa791ca4095dc83bd56d4b534129d8c4abf92cba9682edae8aa24fd5.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "39e8ec776bccd95bd21173fd3380e81bd8e1c246a704d08e8f920224991b4cb1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5ce4fbd1886b01dbceac8019c20689dfe1b2188b6b43bb5df5e70f205caf3c1f", "generation_seconds_batch": 25.15333510900382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.300607204437256, "genuineness": 4.149612903594971, "identity": 0.5866574048995972, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/39e8ec776bccd95bd21173fd3380e81bd8e1c246a704d08e8f920224991b4cb1.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "3a29c99fc70b042eaed811c72cda8872e78b004b6a0435753fe8451fa985dbc6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f567c3e1ab66992cccd812e6427394fd61ce3ca0a9e64e765973d48504fa09e", "generation_seconds_batch": 22.753765064058825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.372091293334961, "genuineness": 4.519899368286133, "identity": 0.4745033085346222, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3a29c99fc70b042eaed811c72cda8872e78b004b6a0435753fe8451fa985dbc6.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": "3a5e43149a76290e684da0425e4410fbe973aee376dfd87f45407085bf3f72c0", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7c39f40ea3aa4314909c84b781cb4ee82eac8198ccb1703e2e18ac6f294f8657", "generation_seconds_batch": 22.583051006076857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 5.95112943649292, "genuineness": 4.193545341491699, "identity": 0.8079847097396851, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3a5e43149a76290e684da0425e4410fbe973aee376dfd87f45407085bf3f72c0.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "3a8effcc64d0d92339126576608661bad6991c60163916def5aaa18d891db092", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe73bff2671c106785be8ff04aa921296ac76640a540fe1f9a26210fd2e2ac13", "generation_seconds_batch": 22.470468753017485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.046441555023193, "genuineness": 4.005263805389404, "identity": 0.6141678094863892, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3a8effcc64d0d92339126576608661bad6991c60163916def5aaa18d891db092.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "3b086ce409374e6c68039338bc11ebb3faaa6b169776b57ed523498774d25568", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0b5ae11fc039511e01aa157a2369c904cef55a3dcc8ef94800f2d62d823b11f", "generation_seconds_batch": 15.416156007093377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3070886135101318, "target_z": 1.5859374630026162, "ce": 6.006728172302246, "genuineness": 4.243753433227539, "identity": 0.47878506779670715, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3b086ce409374e6c68039338bc11ebb3faaa6b169776b57ed523498774d25568.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "3b13a542eb96a3289daed2628b989e2a69b9024105600d501001935b713a571d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ae84d37e063db37441a598021668e86fc1857e650fd2a610c9d5cb087580b3d", "generation_seconds_batch": 21.16075615095906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.046441555023193, "genuineness": 4.607763767242432, "identity": 0.7461729049682617, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3b13a542eb96a3289daed2628b989e2a69b9024105600d501001935b713a571d.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": "3b7a1e68ee692f8235e05225168335ad80b7a0e8d4da1f018d3e7555e1805d6c", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "904f0127abaf29759b8e53976b91a61538b0df50ed0cbd9d69641007c1a26d28", "generation_seconds_batch": 21.180753358989023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 5.990842819213867, "genuineness": 4.118232727050781, "identity": 0.720331072807312, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vaya.", "wer": 0.043478260869565216, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3b7a1e68ee692f8235e05225168335ad80b7a0e8d4da1f018d3e7555e1805d6c.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "3bb615c4709fdbb221ef2a2f5f362b0e154ccfbb9c87513a05c90730f1d268c7", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72b9537ad5e23b424767dc760d88e7dd96595614b510d705bdffa5b2b5d30104", "generation_seconds_batch": 15.459775045048445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6666839122772217, "target_z": 1.1718750632708883, "ce": 6.038498878479004, "genuineness": 4.337893962860107, "identity": 0.6969667673110962, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3bb615c4709fdbb221ef2a2f5f362b0e154ccfbb9c87513a05c90730f1d268c7.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "3bcbe8b4fb66a08e111dc24ae6bd5a231f81d792d2a927d6be78b52db40da674", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44efd4e0dca282ff52b880ac88fc7a9c3d0e2c1292a645a98b31366c451fa5be", "generation_seconds_batch": 21.63251548295375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 6.2767791748046875, "genuineness": 3.8295347690582275, "identity": 0.40365689992904663, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3bcbe8b4fb66a08e111dc24ae6bd5a231f81d792d2a927d6be78b52db40da674.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "23936f3f3ed718e3d64e182ab2a91adaded7095ff732b1b1c73540321ff4cf4d", "generation_seconds_batch": 15.95900203997735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.572874128818512, "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_VOLT_reg_low/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "3c95a42fac2f0a8c82f67ac538f568d66e9de2cb4d487b1f4fc9fa2980904e88", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4be446c1e76be0a3b8158eda63375e94526e0af2311f6561e0575dedbe1de84", "generation_seconds_batch": 16.56943252508063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 5.855817794799805, "genuineness": 4.080576419830322, "identity": 0.8372459411621094, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3c95a42fac2f0a8c82f67ac538f568d66e9de2cb4d487b1f4fc9fa2980904e88.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "3ca90e24a8af64cad823e929838514422415767498f4ddcad4111623d7f2f0aa", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f20fa90f9cb73cd46d92d564e56f527194e799b6f6edd19a7a522f4ff8aa399", "generation_seconds_batch": 14.871567584923469, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.58526611328125, "target_z": 1.2656250353888019, "ce": 6.435632705688477, "genuineness": 4.532451152801514, "identity": 0.4097707271575928, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3ca90e24a8af64cad823e929838514422415767498f4ddcad4111623d7f2f0aa.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "3ccc4e3e05997143bdd095de538ffe0dba15e65624a32751243eaf01006dbda7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59521d327fe76b04627ebb21887400151304a8ec9f71952a9b4bf684be9168ae", "generation_seconds_batch": 16.334758383920416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.103543996810913, "target_z": 1.820312530563056, "ce": 6.229123115539551, "genuineness": 4.193545341491699, "identity": 0.2640770673751831, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3ccc4e3e05997143bdd095de538ffe0dba15e65624a32751243eaf01006dbda7.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "3ce1f6e20b6d2c6e4f9480966b57ba23d3dd81842d28f322f9c645db23ae9803", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "019c1d3d04811dd05d12904916ba4ba7f10a11ede7d1bf9c1dea7160acff1dd7", "generation_seconds_batch": 21.585452035069466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 5.974957466125488, "genuineness": 3.8922953605651855, "identity": 0.519101619720459, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3ce1f6e20b6d2c6e4f9480966b57ba23d3dd81842d28f322f9c645db23ae9803.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "3d280adeb629c8017d48108016c4473ab76f09ad880f752b763c2c456a1e8367", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "599c672b7ab5697df7461d65fb4272b72d8842b27b6559a4c75e9f97da40e324", "generation_seconds_batch": 21.926998381968588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 6.109982967376709, "genuineness": 4.720732688903809, "identity": 0.775022566318512, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3d280adeb629c8017d48108016c4473ab76f09ad880f752b763c2c456a1e8367.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "3d31909d94b0ac2c2f45775a36741cb9c5416ae897da27efd563cdd8bd18c3a2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86921a0ed063a95f6711069cf52e91e2bd6981777a23ec5bfc6c531504485948", "generation_seconds_batch": 22.76585518801585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.181467056274414, "genuineness": 4.017816066741943, "identity": 0.7143521904945374, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3d31909d94b0ac2c2f45775a36741cb9c5416ae897da27efd563cdd8bd18c3a2.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "3d32c3f72b294bb5f5a6d2fc9e88bf948f898f6fe4966a25bbdb021c50d3b79d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8ab027bee89f930392b179b430c1ca49274e58895724c74f8637115c83a405a", "generation_seconds_batch": 23.725946736056358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.173524379730225, "genuineness": 3.8546390533447266, "identity": 0.3421167731285095, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3d32c3f72b294bb5f5a6d2fc9e88bf948f898f6fe4966a25bbdb021c50d3b79d.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "3d9b8c6663a178b4cbac10b3882f312a3cf915b0aee308d7607409b027c1cc5f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "486f736b5263d16e9b7dc06f98cd1b2415d8448b7321d5f6446feb313e37fc61", "generation_seconds_batch": 22.624964650021866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2935189008712769, "target_z": 1.601562549866039, "ce": 6.252951145172119, "genuineness": 4.042920112609863, "identity": 0.4073438346385956, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3d9b8c6663a178b4cbac10b3882f312a3cf915b0aee308d7607409b027c1cc5f.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "3dbaf3fb4c9b15ce0d66471f4801eaa14a8064a618c94d92494ef345a3f3a2d7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d9adeec440597dbfe70a20daabeb533e5655c71c0d64fc866fec0eb7c80a077", "generation_seconds_batch": 16.962318576988764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.103543996810913, "target_z": 1.820312530563056, "ce": 6.292664527893066, "genuineness": 3.9676074981689453, "identity": 0.248526930809021, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3dbaf3fb4c9b15ce0d66471f4801eaa14a8064a618c94d92494ef345a3f3a2d7.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "3dcf9e664acf77e1ec6fcbb19ef7f0199c6b54fbc8c3c2e3366cf5db5618c0c8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95f82df21b4e21b2deb95521f182316ff2a00ec1606fa68f850262f1cdc53f61", "generation_seconds_batch": 20.262846903991885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.2767791748046875, "genuineness": 4.18099308013916, "identity": 0.4547261595726013, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3dcf9e664acf77e1ec6fcbb19ef7f0199c6b54fbc8c3c2e3366cf5db5618c0c8.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "3dd3b6ab745d3d71e1920167c52a6d530e91bbd539575d3d6a1125a6cc92e202", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d330c1fc506d0e5dd6195dd53bbdf472fad02675ffd059f0de445325947152c", "generation_seconds_batch": 21.730016452958807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 5.855817794799805, "genuineness": 3.8295347690582275, "identity": 0.5415034294128418, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3dd3b6ab745d3d71e1920167c52a6d530e91bbd539575d3d6a1125a6cc92e202.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "3e0bced9533c9d1c044e6036bcb73bb5c2b10ce844672da1e8e1662fea97f017", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab5d52336407b43d7626d908db4f7037ccef0ccb35de7354daa96dbc8a86812b", "generation_seconds_batch": 15.883716267999262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.673468828201294, "target_z": 1.1640624512063489, "ce": 6.117925643920898, "genuineness": 4.406930446624756, "identity": 0.7380861639976501, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3e0bced9533c9d1c044e6036bcb73bb5c2b10ce844672da1e8e1662fea97f017.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "3e88ae88f9bc9160b87881925289830bd1e0c80153dffe1c90c78bae177e5711", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6deb83f833c31e96eda2a02cd6fe1c47ad32866f953d83c28c037b463bdd6ba2", "generation_seconds_batch": 23.87584584497381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5132648944854736, "target_z": 2.499999897050758, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.6591287851333618, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3e88ae88f9bc9160b87881925289830bd1e0c80153dffe1c90c78bae177e5711.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "3f17359629913e225e5cc57b345a321dd881679617d70247faa3f47d112585bd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33d162579ba5a60d9a8e7aa19688631e42ac1915d9afb71e6699653e9524cb86", "generation_seconds_batch": 20.69444750796538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.284721851348877, "genuineness": 4.293961524963379, "identity": 0.6867128610610962, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3f17359629913e225e5cc57b345a321dd881679617d70247faa3f47d112585bd.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "3f52f4493667c426be940af546aa0fa191e3d00a6e2aaee3f239f5205b200dd2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d92a05dc1b33141137b51f978bc061c660f33eb9151b2b45621ea09a135c6a4f", "generation_seconds_batch": 22.72625246993266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.181467056274414, "genuineness": 4.6203155517578125, "identity": 0.8567720651626587, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3f52f4493667c426be940af546aa0fa191e3d00a6e2aaee3f239f5205b200dd2.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "3f9c85228464ee2aab2989690f9a8c32e5ec1c1cdae030526937d9985c16f63d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69e603410a4f131e06dbc2623e69d8c861fe4c1c05fbdf6c81650c93b86e4489", "generation_seconds_batch": 23.21691831899807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.149696350097656, "genuineness": 3.998987913131714, "identity": 0.4711223840713501, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/3f9c85228464ee2aab2989690f9a8c32e5ec1c1cdae030526937d9985c16f63d.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": "3feabf3dca111faea220f644e43c1d9e14d6e044cbcbce79d48b0de1245c789e", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bfa38d5b1ce9619864d8534c51595670576e323f14fa89107d825df55f6752c", "generation_seconds_batch": 15.875433427048847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.292664527893066, "genuineness": 4.507347106933594, "identity": 0.6536294221878052, "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_VOLT_reg_low/3feabf3dca111faea220f644e43c1d9e14d6e044cbcbce79d48b0de1245c789e.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "405f3fccb77046f6e7202533e00b5c02951b9bc26ffc4a3cb575ec8d2662a47a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9477e1866613393c0a40a14a2d90abf4dd391457531e22db5af2a1be138a2cf", "generation_seconds_batch": 16.06397369899787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.284721851348877, "genuineness": 4.369274139404297, "identity": 0.6548408269882202, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/405f3fccb77046f6e7202533e00b5c02951b9bc26ffc4a3cb575ec8d2662a47a.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "40c792f38bd11fed03adea8eab3bf380b5685ec324acda879d0b7448bb88cf78", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90d9b4bc0a0d7dd94200453ea62ba8ffa02560af4fc4763005c135bf4a418f0d", "generation_seconds_batch": 23.503076920984313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.41180419921875, "genuineness": 4.243753433227539, "identity": 0.6145346164703369, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinkommen.", "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_VOLT_reg_low/40c792f38bd11fed03adea8eab3bf380b5685ec324acda879d0b7448bb88cf78.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "40dbfb91bc70e62b2086107804d5b321cd1aae9a6a4ef4044d64370f731e52e9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85dd354d9f57582a3e71bbd2678e224187a5895a54e4b39d2cc61f5bc5ee3291", "generation_seconds_batch": 17.26126123801805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 5.935244083404541, "genuineness": 3.835810661315918, "identity": 0.6306271553039551, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/40dbfb91bc70e62b2086107804d5b321cd1aae9a6a4ef4044d64370f731e52e9.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "4160c4d2c608b55e7e09d54aff6ea4ddfa417b4f506541eac1075abead20fd50", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54408cf3b175d1aab90819f48ffc09c288403de50bf5623f0a9138353233dbdb", "generation_seconds_batch": 20.96434709301684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.38797664642334, "genuineness": 4.268857479095459, "identity": 0.5183193683624268, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4160c4d2c608b55e7e09d54aff6ea4ddfa417b4f506541eac1075abead20fd50.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "419f89f8060f4e13171fa3ff19bbeeeb1a7ebc9fdcd458b3ac980998edc4b01b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6fa8b412a642ab249768785626603e7d82c1e16af681ae5c33a40e91e72158b", "generation_seconds_batch": 22.176205517025664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.109982967376709, "genuineness": 4.105680465698242, "identity": 0.28100937604904175, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/419f89f8060f4e13171fa3ff19bbeeeb1a7ebc9fdcd458b3ac980998edc4b01b.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "41d094d3e712a67adffd5f40517c5559267376138c7b2bc516e8f0e1c24197e6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ad7e7c65fe852dfde9bbef2e99e7075a95128cf9c6d5e8b9979f8aa14053c2b", "generation_seconds_batch": 22.97760985395871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4224305152893066, "target_z": 1.4531249796246293, "ce": 6.221180438995361, "genuineness": 4.494794845581055, "identity": 0.7216596007347107, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/41d094d3e712a67adffd5f40517c5559267376138c7b2bc516e8f0e1c24197e6.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "420123d2d1bf0e63be8e1e1c963d52f60803193f0a95922724d95a51f86c9809", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8758212ab387275378fda64c03246fad2ddd7aa662e02939f84bf8e3b018f13", "generation_seconds_batch": 23.02227972599212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7303789854049683, "target_z": 2.250000017158207, "ce": 6.348263263702393, "genuineness": 4.281409740447998, "identity": 0.8385428190231323, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/420123d2d1bf0e63be8e1e1c963d52f60803193f0a95922724d95a51f86c9809.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "421f3efd46047fea7cfb345d9963216b019ee3057babf36bc01cdef414276fc6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45bb5628db70ed1d695bc6b5786ce0616b1cd00c464549e95ad8009fc04bdbe9", "generation_seconds_batch": 14.1354167109821, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.6592975854873657, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/421f3efd46047fea7cfb345d9963216b019ee3057babf36bc01cdef414276fc6.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "425976609ad0ae3c71f5d9bcf53f7e65d307fefba49f1dd7c22d7e7b47e99b5b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2b1535444f837d80773312842b5cf5fb1a36dc69459e539e0d6b97649d90822", "generation_seconds_batch": 22.95010663208086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3817216157913208, "target_z": 1.499999965683586, "ce": 6.070269584655762, "genuineness": 4.20609712600708, "identity": 0.7181931734085083, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/425976609ad0ae3c71f5d9bcf53f7e65d307fefba49f1dd7c22d7e7b47e99b5b.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "42c6e691c1620bf12964c805ec48870b453e003eeeda3fd69644f2de0bf8c216", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50a9ae3d1dd0a1a2bacc9034d08f83701b3161f91e80e965803ed1fc94690cc6", "generation_seconds_batch": 16.463833378977142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.476709008216858, "target_z": 1.3906250439679053, "ce": 6.221180438995361, "genuineness": 4.425758361816406, "identity": 0.6269857883453369, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/42c6e691c1620bf12964c805ec48870b453e003eeeda3fd69644f2de0bf8c216.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "43416c67e2f7597c69e037c4c981b116f7cfcde6efcb49d4db09e1097f85344c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99944b293ece0d90f7ec42225774284f6e965f9266eb08c7fa5bd3a535d3364c", "generation_seconds_batch": 22.238804621971212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1713922023773193, "target_z": 1.7421875080429094, "ce": 6.197352409362793, "genuineness": 4.65797233581543, "identity": 0.7352238297462463, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/43416c67e2f7597c69e037c4c981b116f7cfcde6efcb49d4db09e1097f85344c.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "7e74883e25d2083d18d34cb9c316cb307cc4a0712e4eb4ff56d9c207efddc0e3", "generation_seconds_batch": 18.00348633399699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.917722225189209, "target_z": 0.8828125348526079, "ce": 6.141753673553467, "genuineness": 4.394378662109375, "identity": 0.7237043380737305, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "4367927f791cc554342a6b9e6dbef42e6c2d8a61491e7030358286723af8e88c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8c63d5920620ba414e3310277329c08dcf84c922a872e747f7bcb1ac8c81487", "generation_seconds_batch": 22.6369284739485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 5.919358730316162, "genuineness": 3.8169827461242676, "identity": 0.7330641746520996, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4367927f791cc554342a6b9e6dbef42e6c2d8a61491e7030358286723af8e88c.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "43933484203c6756695af4cad89d7ecc7ff311b781bcd28c846b4cd1f28f75c6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3326bed36e8f4e48ac2b1a566229f84d742bbe7e401018e7dfd599bed025a8c9", "generation_seconds_batch": 20.600569118047133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4292153120040894, "target_z": 1.4453125048257458, "ce": 6.149696350097656, "genuineness": 4.432034492492676, "identity": 0.7759150266647339, "hypothesis": "La porte 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_VOLT_reg_low/43933484203c6756695af4cad89d7ecc7ff311b781bcd28c846b4cd1f28f75c6.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "439aa22cd7b4c7b797f136d1e6d9c882cd44ff38d060eb8620966c29c9bdd696", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c411ffd1627d5b7698cc50a922c8e0553e41b4df3cbebddf609aababe8bd1450", "generation_seconds_batch": 22.66318360203877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.133810997009277, "genuineness": 4.394378662109375, "identity": 0.6829549074172974, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/439aa22cd7b4c7b797f136d1e6d9c882cd44ff38d060eb8620966c29c9bdd696.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "52d92d4db811c698036563ed501c9a3b7eca864a0c4e110de4907fe3aa964063", "generation_seconds_batch": 18.816525902948342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.260893821716309, "genuineness": 4.331617832183838, "identity": 0.5532110929489136, "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_VOLT_reg_low/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "44c60b5f49473cf00c09a6ca67b596ac29c191588ed3a1f8f9e60bd40fa97aaf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c695104b98d53bb886c29dc8a1ae40c3b7f524599b128f929c59285e8678941f", "generation_seconds_batch": 16.676845018053427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.300607204437256, "genuineness": 4.055472373962402, "identity": 0.5124928951263428, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/44c60b5f49473cf00c09a6ca67b596ac29c191588ed3a1f8f9e60bd40fa97aaf.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "44e67d0aa5bd3401dbf9fc0bd898c6083923d63563f810d7b0be58957bbc1754", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e41faca0c4b3409e8fd1dfbda72d19ad38e3221289cad3e874b0fc6876bff7d", "generation_seconds_batch": 16.09410828293767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 5.959072113037109, "genuineness": 4.093128204345703, "identity": 0.45109131932258606, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/44e67d0aa5bd3401dbf9fc0bd898c6083923d63563f810d7b0be58957bbc1754.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "459310957eacb954ad081a813d1a34bd11dd2aff9068a532960dd13d73576b08", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6037f19ed7df1507a6b5d56ed2fde49cd6ddd018fc623ac9ac90a32654c7e771", "generation_seconds_batch": 15.761161356931552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.6803929805755615, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/459310957eacb954ad081a813d1a34bd11dd2aff9068a532960dd13d73576b08.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "ec9bc7abbc7aeef333126825b0fdb9043b8fb51a5999bf65bd47b4252a0cc9f3", "generation_seconds_batch": 16.076739610056393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.252810001373291, "target_z": 1.6484375359249959, "ce": 6.403861999511719, "genuineness": 3.898571252822876, "identity": 0.5023207664489746, "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_VOLT_reg_low/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "4612f5ad07a4e8060b0954ac8198a3f51cfaae426013da20c828b36b632b9196", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6de848db0e266cd6c2af109e76ba6dc3fca6eedaf2ef6cd1de2b28e166e6126c", "generation_seconds_batch": 13.997606182005256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.165581703186035, "genuineness": 4.306513786315918, "identity": 0.4939504563808441, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4612f5ad07a4e8060b0954ac8198a3f51cfaae426013da20c828b36b632b9196.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "463870588e4f441a527ced8a2295c38c594dfce581812331f84526232fc59919", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f87c9f8f31b2c07fd2787cebea3c251a17c88f5a7a3717122c575a178ba8e8e5", "generation_seconds_batch": 22.50529301003553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 5.808161735534668, "genuineness": 3.7416701316833496, "identity": 0.6620585918426514, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/463870588e4f441a527ced8a2295c38c594dfce581812331f84526232fc59919.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "466c700ae05b9b2a8a91606432fc6d7a7d30b159d99f419dd863d19da4af4a68", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d6fef715eb1be0dad3c56fa23b4e08c814ca44aff5fe5f9144d17c0fd0e678ff", "generation_seconds_batch": 15.033456400036812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1713922023773193, "target_z": 1.7421875080429094, "ce": 6.260893821716309, "genuineness": 4.720732688903809, "identity": 0.6804643869400024, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/466c700ae05b9b2a8a91606432fc6d7a7d30b159d99f419dd863d19da4af4a68.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "4689e54b4e8ce07deb5514c5b0238f93c4e2f53c46480e70d751aa981705bd82", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb4325dd02cdbc2474f82315a248f70a3c8c4ff2e90fa23c47f8a193fcd07504", "generation_seconds_batch": 15.520435606013052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7616714239120483, "target_z": 1.0625000042895518, "ce": 6.268836498260498, "genuineness": 4.607763767242432, "identity": 0.7999004125595093, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4689e54b4e8ce07deb5514c5b0238f93c4e2f53c46480e70d751aa981705bd82.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "46f46be74f539a6304a55020249bb7fd630f7267da00f19cff4b8ed0c8024461", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "534ba4d06235ea8f82b1bafc6b77d222a7972c50d68937c104fe963a30ecef14", "generation_seconds_batch": 18.92224186391104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 5.649308204650879, "genuineness": 4.444586753845215, "identity": 0.8571689128875732, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/46f46be74f539a6304a55020249bb7fd630f7267da00f19cff4b8ed0c8024461.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": "4755d5aa45cb7b8b78649ff9d73c9ba7d56b0c73347a504b1a7d63ee14466874", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e5f5ad23c7b83fb9efcb12b7a81a8cfc37c46c12f607e88eaeed1e2e1301ce1f", "generation_seconds_batch": 15.206300494959578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.308549880981445, "genuineness": 4.325342178344727, "identity": 0.5326768755912781, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4755d5aa45cb7b8b78649ff9d73c9ba7d56b0c73347a504b1a7d63ee14466874.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "536f6c941e8f4b1eaa75218801ad9d31d42ef612a162fd808cbd4c1f8a38e21c", "generation_seconds_batch": 15.187313472968526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.7974070906639099, "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_VOLT_reg_low/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "47f07f8e1dd07c11e22eb5833fe2073184a23e32fe5b1540966696acc7ccf2bf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 131, "seconds_generated": 10.48, "mp3_sha256": "8fdaad3852fe06fc732015d94a13bbf5fcad33141d28f08c9bfb73a11615d91a", "generation_seconds_batch": 24.254362634033896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7303789854049683, "target_z": 2.250000017158207, "ce": 5.816103935241699, "genuineness": 4.319066047668457, "identity": 0.8269132375717163, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce cas.", "wer": 0.10526315789473684, "duration_error_s": 0.28000000000000114, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/47f07f8e1dd07c11e22eb5833fe2073184a23e32fe5b1540966696acc7ccf2bf.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "481321c8bdd7288e6b0aaf4b6319048c16b51e239404d985779ffe963c8f77b7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3695380193fbf69dc0857aebc1e090da28be8d3c06f07fb6a0f8e1398f2d8ed6", "generation_seconds_batch": 20.618647772003897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.229123115539551, "genuineness": 4.2751336097717285, "identity": 0.2641139626502991, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/481321c8bdd7288e6b0aaf4b6319048c16b51e239404d985779ffe963c8f77b7.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "48573fe8dc55892037426e615d534b631c867ee26fc3424f4f84bd66512b6f8d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a7359887f675af3c23aa84b640b7f64be71c34e93f87d52b9a8211dd30fec436", "generation_seconds_batch": 21.62753762502689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 5.927301406860352, "genuineness": 4.093128204345703, "identity": 0.6149322986602783, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/48573fe8dc55892037426e615d534b631c867ee26fc3424f4f84bd66512b6f8d.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "488551e1f66365099b23603353c39002781156c3e9cb70dbcaf0bffdfcc1cc3e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29d73cba420b4f2ff56a27caeb698c9e7c04b4c058658502c2f5889b6f840a6f", "generation_seconds_batch": 22.29804027301725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.117925643920898, "genuineness": 4.162164688110352, "identity": 0.6223476529121399, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/488551e1f66365099b23603353c39002781156c3e9cb70dbcaf0bffdfcc1cc3e.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "48daac38afff864ebeb690fc3e97f7ed4c26421aab1e6b7f57ffddc0555fa61b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0afeacba4de7bbc10ce8409934adf5a66ba3498fbf0c38e3c2315e9f9bd3ee67", "generation_seconds_batch": 15.140744079020806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 5.744620323181152, "genuineness": 4.3127899169921875, "identity": 0.8400790095329285, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle connaît.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/48daac38afff864ebeb690fc3e97f7ed4c26421aab1e6b7f57ffddc0555fa61b.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "498145e7ffa77232be52f1d5de2179b2f136ecccebdb58d32e6ea5c9357e995b", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d8337ba00bddb583c5e6964eae457f07f83aedb8e3ccbd27ad2ebba6913f424", "generation_seconds_batch": 17.199576804996468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.316492557525635, "genuineness": 3.7165660858154297, "identity": 0.46400177478790283, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/498145e7ffa77232be52f1d5de2179b2f136ecccebdb58d32e6ea5c9357e995b.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "49ac74caa0528bb174cfc7eacae94b9da58febba448cf2f1a9ba9d0e060f8dd0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6e7b822c528bc31bf66cb7772da99ec98894bbbda25afabe46c697a814f7ed93", "generation_seconds_batch": 20.71053799102083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.427689552307129, "genuineness": 4.250029563903809, "identity": 0.6391037702560425, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/49ac74caa0528bb174cfc7eacae94b9da58febba448cf2f1a9ba9d0e060f8dd0.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0982e640ec694cc1df8d141ed5c4ca4d3af8a2455cc115638f0328c1e6e31c6b", "generation_seconds_batch": 15.909555285004899, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.332377910614014, "genuineness": 4.494794845581055, "identity": 0.6667637825012207, "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_VOLT_reg_low/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.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": "4a3662be42b79ac4ffebc53f0daa4a842dd2ab3251d111baf292566ce37e4051", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11bdb0ff80ba1f4c4a1c754a14fa06d7b01e6c2ad479daf74bd4040fc7fe4598", "generation_seconds_batch": 24.11141682509333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 5.879645347595215, "genuineness": 4.432034492492676, "identity": 0.712358832359314, "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_VOLT_reg_low/4a3662be42b79ac4ffebc53f0daa4a842dd2ab3251d111baf292566ce37e4051.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": "4a5a045606dbcefbdb0650d4ab782a44574879ee88f99ffc19ee964de752e487", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b43e11768f6c75cce4a0fcf0acd795cb858a41c2ff4770ab8b31bdd061b2eb1", "generation_seconds_batch": 17.261192742967978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.3641486167907715, "genuineness": 4.306513786315918, "identity": 0.668869137763977, "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_VOLT_reg_low/4a5a045606dbcefbdb0650d4ab782a44574879ee88f99ffc19ee964de752e487.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": "4a8ce1449e79a578ac322c7a16f13089c6b9f134d74050cee2c8dedb8285a6ea", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd02e70609143c4e7b8b5d55dd530a94c7ceebcf8dc1d5bfcb7bff1f210d7d14", "generation_seconds_batch": 22.300867998041213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.562715530395508, "genuineness": 4.331617832183838, "identity": 0.16135719418525696, "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_VOLT_reg_low/4a8ce1449e79a578ac322c7a16f13089c6b9f134d74050cee2c8dedb8285a6ea.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "4aa27e6ca143f58d4f6c21c03b23d0dd0b7a48fe20b14015ddebac5ab9205755", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73a6ae82182a59839b5507cd336b3b35e56871f74e4c2bb41ff5708be831a096", "generation_seconds_batch": 21.781117214006372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.2767791748046875, "genuineness": 4.281409740447998, "identity": 0.71207195520401, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinführen.", "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_VOLT_reg_low/4aa27e6ca143f58d4f6c21c03b23d0dd0b7a48fe20b14015ddebac5ab9205755.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "4ad958cc9cb482f022b8d91b83e0124b72095fb377a50972adde535bf2047013", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "48a03112ef96d5f12426a5d171db20902ca1ac0169857bc07d2c8d8606306868", "generation_seconds_batch": 17.756354300072417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.260893821716309, "genuineness": 4.250029563903809, "identity": 0.6525512933731079, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereitst.", "wer": 0.10526315789473684, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4ad958cc9cb482f022b8d91b83e0124b72095fb377a50972adde535bf2047013.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "4b197ee02d4ebce07b2cc4d85ed05b10c613c13fcfb92eb14544cf904cd0b6ed", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c903700289487f71e180aa05e0a24d65126095fa24559f5d7f80ecbaf5641d73", "generation_seconds_batch": 20.78895715798717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.133810997009277, "genuineness": 4.645420074462891, "identity": 0.8718247413635254, "hypothesis": "La porte 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_VOLT_reg_low/4b197ee02d4ebce07b2cc4d85ed05b10c613c13fcfb92eb14544cf904cd0b6ed.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": "4b9e513a44f8d66229741ddbb7ba5ab96ab751a1077733eeb3417d3083f1f87a", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a76bdb03b2c3ea7cc5bd363eb62edd61a0257773d7bc80b841a840f672afaac3", "generation_seconds_batch": 16.07816788495984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.1020402908325195, "genuineness": 4.413206577301025, "identity": 0.9132102727890015, "hypothesis": "La route tourna gauche après le pont, continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4b9e513a44f8d66229741ddbb7ba5ab96ab751a1077733eeb3417d3083f1f87a.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": "4bd4e4f4cca2f226ea9347ce057573761b1602483f366c5b0e3dd096a19a9d0d", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8a20a6ee042dfe91f7ec76625abb6ab4cc18ef9770b3e9a017aeb208eb4b89e0", "generation_seconds_batch": 15.437324968981557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.117925643920898, "genuineness": 4.268857479095459, "identity": 0.7633119225502014, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4bd4e4f4cca2f226ea9347ce057573761b1602483f366c5b0e3dd096a19a9d0d.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "4c174dfb33e13af3ab754d08029ad58a0b96b2bb3b8fd1f7a745fde0d0b94ee1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "306be687926c3b7f8d795ce2b820ce1a96e9d71ead66769973860d26cef92eaf", "generation_seconds_batch": 17.989715995034203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.6481053233146667, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4c174dfb33e13af3ab754d08029ad58a0b96b2bb3b8fd1f7a745fde0d0b94ee1.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "4c3a778e56d6caac32899347db4b923ab3df700c3780829abd73848d4094eff4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "c57e21ba6957726244992a80bc6ccc116ead2c6620a54d4aebeea594658e45ff", "generation_seconds_batch": 21.05792363290675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 5.967014789581299, "genuineness": 4.6203155517578125, "identity": 0.7166558504104614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle a.", "wer": 0.05263157894736842, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4c3a778e56d6caac32899347db4b923ab3df700c3780829abd73848d4094eff4.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "4d0a04e9f37b3344735ccb179493ea8b052dc2db692ba330f69a6188074afeb2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d134696c94eda046784e17029df51e741fef81e8a91be058dffac8a0eddfa10f", "generation_seconds_batch": 21.414761896943673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0764046907424927, "target_z": 1.851562567024246, "ce": 6.149696350097656, "genuineness": 3.986435890197754, "identity": 0.4139878749847412, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4d0a04e9f37b3344735ccb179493ea8b052dc2db692ba330f69a6188074afeb2.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "4d29a9014ea26b54cc721bc3dfee3040650aca68d6704b80db7cc1ecf4ac7f92", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dd9cb6ab4b3267388fde9171c9935fdbfdddd5e3c1d9af9df99e93aa87fa2732", "generation_seconds_batch": 16.78196852398105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6666839122772217, "target_z": 1.1718750632708883, "ce": 6.300607204437256, "genuineness": 4.105680465698242, "identity": 0.665925145149231, "hypothesis": "De deur 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_VOLT_reg_low/4d29a9014ea26b54cc721bc3dfee3040650aca68d6704b80db7cc1ecf4ac7f92.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": "4e125d05c3ade4a0a99d1f771edea7ff610d85120ccab07849ef8803c79df516", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8669c45ad3183ac08fca92ea60b3b843bf96eff107c21e3cafcb9aad2ef28693", "generation_seconds_batch": 21.494886144064367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.340320587158203, "genuineness": 4.432034492492676, "identity": 0.7578712105751038, "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_VOLT_reg_low/4e125d05c3ade4a0a99d1f771edea7ff610d85120ccab07849ef8803c79df516.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": "4e2e2a46046ff6895a40661ef1e306eb9ed80a162ec77ab345b211ddfd857b06", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e7564debb071aa8c2134fea5140b89cd8d1f7f2aad087feb1950bfbcc95c0b3", "generation_seconds_batch": 25.238112117978744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.078212261199951, "genuineness": 4.218649387359619, "identity": 0.7919453382492065, "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_VOLT_reg_low/4e2e2a46046ff6895a40661ef1e306eb9ed80a162ec77ab345b211ddfd857b06.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "4ee126900af392c781bee10927ab59644b0bc3e84b76c80b2cc76342fa38465e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f930add68ae94dfdf1ee56c96fddb8ef741135a2ad2d43255e91f9c0828e7a62", "generation_seconds_batch": 16.420651596970856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.332377910614014, "genuineness": 4.030367851257324, "identity": 0.2861706614494324, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4ee126900af392c781bee10927ab59644b0bc3e84b76c80b2cc76342fa38465e.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "4f14e6dc383102dff6ace10330d20bb5fa37a9f6bea572a47b1555551c453fce", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f724bc7c3b4672c1ae27501067ba43d5487fc9b65fdf5d7cfc6c23262938781", "generation_seconds_batch": 28.309883030015044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.2767791748046875, "genuineness": 4.337893962860107, "identity": 0.6964913606643677, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4f14e6dc383102dff6ace10330d20bb5fa37a9f6bea572a47b1555551c453fce.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "4f1f6ebb6f53d3f00824210315779a269e8fef892c3fc03dbfd4ac234ad14ea2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "55c25566237fa75431727420667079da8822b636dfbddc62d97f6a27fc024aa8", "generation_seconds_batch": 23.715554426889867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.5831769704818726, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreiben.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4f1f6ebb6f53d3f00824210315779a269e8fef892c3fc03dbfd4ac234ad14ea2.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "4f3559d9842f82d9aa1f006003501027fe8d5601e901f343e87ac93df7ecfd14", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "166c8868923faa6a132a0888504fb4cf2a3886a5e2d9711005a4f8e5cd4060eb", "generation_seconds_batch": 19.696258637937717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.268836498260498, "genuineness": 4.268857479095459, "identity": 0.7556684017181396, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4f3559d9842f82d9aa1f006003501027fe8d5601e901f343e87ac93df7ecfd14.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "4f4d8a8da6bab121af2121072c85f0662a7fcaf8c70732a26fbbb551e496dea3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a196e9031d29f0bf1975806926bd79b1e68c58f350097583935ed14d2e73781f", "generation_seconds_batch": 21.035757326055318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.260893821716309, "genuineness": 3.873466968536377, "identity": 0.40887707471847534, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4f4d8a8da6bab121af2121072c85f0662a7fcaf8c70732a26fbbb551e496dea3.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "4fb2b7681c6f2add096cfd5c74af1d255b7f50fb35aee7a960b681d670955d00", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fa8e0ef56c8222cfaa0936444c0401d1261e681f8548fe52d7a896d047644b2e", "generation_seconds_batch": 22.17207066901028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.395919322967529, "genuineness": 4.118232727050781, "identity": 0.6386955976486206, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4fb2b7681c6f2add096cfd5c74af1d255b7f50fb35aee7a960b681d670955d00.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "4ff6323258e19f160382fd20096e0f0ff41d2ac4dba9920a70b3631bc14e3be5", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b53e535c8a9d464e7f0fe62fb794cbf86a93484913795fe81d90a7e75fc6f79", "generation_seconds_batch": 15.459775045048445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5038483142852783, "target_z": 1.3593750075067155, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.7216533422470093, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/4ff6323258e19f160382fd20096e0f0ff41d2ac4dba9920a70b3631bc14e3be5.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "50c422153e9eb1e4f9fcf3476ff84dbe0566408bb7b4adf9884d1b9bdcc12027", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ea820d1a173f014b3931e5d05555baf96652a873a03775f804ccf19bda3f92f", "generation_seconds_batch": 22.37740361096803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 5.927301406860352, "genuineness": 3.9927117824554443, "identity": 0.6719483137130737, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/50c422153e9eb1e4f9fcf3476ff84dbe0566408bb7b4adf9884d1b9bdcc12027.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "510ab172e4624fd341ffaef5fd8e74014201a4fc577f99d0ca20f85a004de26a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7e7b49bb5406d4eeae3fd628190798d9de7afd512e0c34d606c8d2d9cca0167", "generation_seconds_batch": 22.238804621971212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.205295085906982, "genuineness": 4.413206577301025, "identity": 0.5697585940361023, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/510ab172e4624fd341ffaef5fd8e74014201a4fc577f99d0ca20f85a004de26a.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "517b9a0e3c9cc9799f519799ae1e8a9f3014c9105283e06a28f8e0545a0284b3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf3fd448687c4d0ce21bc52c9f2dab86bb43c38c4850d8d56339e1455a7955d6", "generation_seconds_batch": 20.229721048963256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 5.919358730316162, "genuineness": 4.030367851257324, "identity": 0.729282557964325, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/517b9a0e3c9cc9799f519799ae1e8a9f3014c9105283e06a28f8e0545a0284b3.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "51aa8a85947bb4d5cc676b8efde907a66c5cb5d3705abf92413f7f3f47ff4630", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1e7cebdb6916e2202df17f99f568765eb3e8f94f15847df419ff53aa3a406f8", "generation_seconds_batch": 17.41362712800037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.229123115539551, "genuineness": 4.061748504638672, "identity": 0.23953765630722046, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/51aa8a85947bb4d5cc676b8efde907a66c5cb5d3705abf92413f7f3f47ff4630.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "51b2f246604cb3191288a03163b3254ab20c668d4c6a7ee777ef68b7adea50eb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "710c8eaeea78870844164d3c5c983baf36b391ca21745ef6153f57bcada6ea02", "generation_seconds_batch": 19.855579255963676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.266379714012146, "target_z": 1.632812449061573, "ce": 6.229123115539551, "genuineness": 4.2249250411987305, "identity": 0.3570946455001831, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/51b2f246604cb3191288a03163b3254ab20c668d4c6a7ee777ef68b7adea50eb.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "522ce46c2dc96dc7b7ae83ec6a878dfe77892cecbd377a85b0ac0ffa029e362c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d0b0e190533e174ee5f6adc54bafb93c776dd7f3bf91dd24660beda3d8755f04", "generation_seconds_batch": 23.912634007050656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.538887023925781, "genuineness": 4.469690799713135, "identity": 0.7115317583084106, "hypothesis": "Ik 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_VOLT_reg_low/522ce46c2dc96dc7b7ae83ec6a878dfe77892cecbd377a85b0ac0ffa029e362c.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "5247445f463e2cdcbf3fed8e3b2183a3268b7b38d85f545268ac663e8a59382f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "270ac2049a918e19ba860b73b6c0ce9dae8bdb46bb7b9b4fb8c8abef5075d964", "generation_seconds_batch": 19.56816331099253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.1894097328186035, "genuineness": 4.105680465698242, "identity": 0.790769636631012, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5247445f463e2cdcbf3fed8e3b2183a3268b7b38d85f545268ac663e8a59382f.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "52db836c820059b258ae160eefa84776e98b0303b9ce9356c915c1c3d95156c2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "94594aaf3fbdd8abc4db0ef722aed682f64a9e4e0b4e30ce107a87380527ba00", "generation_seconds_batch": 16.697846231982112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.403861999511719, "genuineness": 4.519899368286133, "identity": 0.26728057861328125, "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_VOLT_reg_low/52db836c820059b258ae160eefa84776e98b0303b9ce9356c915c1c3d95156c2.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": "535ad7617f57062c7d0ed4cb4d67f13defa5dfb041ec4e7fd8ff6ec798ae8444", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da62c28e6ed44dc9dcb50be8ff8164a231016096ba0005e2274dbeb2357ebde8", "generation_seconds_batch": 14.38022274908144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.38003396987915, "genuineness": 4.20609712600708, "identity": 0.3078383207321167, "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_VOLT_reg_low/535ad7617f57062c7d0ed4cb4d67f13defa5dfb041ec4e7fd8ff6ec798ae8444.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "53a96ca982d7a7e9f978e0f05dba85fa1411cd5014a268134a45d464f320e3b2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2321da239095deccc2fc78ffe050381d77eaeed41203da087e559fda9e4ff3b9", "generation_seconds_batch": 22.9904407360591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.205295085906982, "genuineness": 4.444586753845215, "identity": 0.8953642249107361, "hypothesis": "Ik 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_VOLT_reg_low/53a96ca982d7a7e9f978e0f05dba85fa1411cd5014a268134a45d464f320e3b2.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "53b6de70b8fa6c172830f79d5adbdb73bbca36043188694177c15b3e2671ae4e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c970f6e3dc8985a3cb4c6b0786240c829c5c29337d432b686303182c777d0af4", "generation_seconds_batch": 15.320706845959648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.499174118041992, "genuineness": 4.15588903427124, "identity": 0.6063659191131592, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/53b6de70b8fa6c172830f79d5adbdb73bbca36043188694177c15b3e2671ae4e.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "07263fb83454d137d893ec54bf65cbb615eadbbbbde688132cbd4aaed18e111f", "generation_seconds_batch": 16.076739610056393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.300607204437256, "genuineness": 3.8922953605651855, "identity": 0.5177460312843323, "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_VOLT_reg_low/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.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": "5499c8cf65ff410baa0e4f85a52969187e6726c017dec4ed958c2de2eec41ea4", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ec23c9965afd57735ff2c71aa42d3817efe497f78b08847903fecfabb07b1b82", "generation_seconds_batch": 14.844497876008973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.149696350097656, "genuineness": 4.582659721374512, "identity": 0.8149155378341675, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5499c8cf65ff410baa0e4f85a52969187e6726c017dec4ed958c2de2eec41ea4.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "54cd895e47a12ec5d0f6ac6df23d2b702717a2cca67b6afa2d77b4fd1bda9a1a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "073802aa4c97688d23512317d5ee3e0a587acf49e609b7d98623ddd25e59a2be", "generation_seconds_batch": 18.894496402004734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 5.982900142669678, "genuineness": 3.948779582977295, "identity": 0.5435954332351685, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/54cd895e47a12ec5d0f6ac6df23d2b702717a2cca67b6afa2d77b4fd1bda9a1a.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "54ef6f6afb5d05ff8ba00a61a6f623b9e20b896ad4d4609bfd5c12ab86d494ff", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7ec2be5c553632aa9a5a665b4ae0e666b89673d77f03831f4e800e469b109bd", "generation_seconds_batch": 16.138175346073695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6802536249160767, "target_z": 1.1562499764074654, "ce": 5.935244083404541, "genuineness": 4.570107460021973, "identity": 0.676557183265686, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/54ef6f6afb5d05ff8ba00a61a6f623b9e20b896ad4d4609bfd5c12ab86d494ff.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": "5553acb23b9490248eef8c5d8752200f1e3dde385defa35e5f4ce3a4bc96e311", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e5d04b2369b859578245542d9055324325b8b91855cee11a952b0d0fadc6a3b", "generation_seconds_batch": 16.431593523011543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 5.855817794799805, "genuineness": 4.231201171875, "identity": 0.757138192653656, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras in a 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_VOLT_reg_low/5553acb23b9490248eef8c5d8752200f1e3dde385defa35e5f4ce3a4bc96e311.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "557600629a419cbbc161817b96734426fb87c5b2e9906ebb3b6a9056e3813b11", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4242fdc6da9c1019fadcaa8913c5692b6f1291f9a910945712c11b99db92e59", "generation_seconds_batch": 21.3136409180006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.395919322967529, "genuineness": 3.823258876800537, "identity": 0.5474547147750854, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/557600629a419cbbc161817b96734426fb87c5b2e9906ebb3b6a9056e3813b11.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "559e0ef1eb952bc3a570dcb49b968fccc1b67b46e8db5c9d79e4518bee532dd2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cb4736b5fdcf078685896064abd5870a5bf78cfb46b56a6d943bc9561cc813c", "generation_seconds_batch": 21.61740483203903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.340320587158203, "genuineness": 4.218649387359619, "identity": 0.5706494450569153, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/559e0ef1eb952bc3a570dcb49b968fccc1b67b46e8db5c9d79e4518bee532dd2.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "560f5ab291f719dd00239232a6a745365c87f956e59223745602b7433129475c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "716b206fd31d946963b1bdfb167fa037b7faa97e8ad20ae9fe48b8494ebe2f65", "generation_seconds_batch": 17.17496317392215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6531143188476562, "target_z": 1.1875000128686553, "ce": 6.284721851348877, "genuineness": 4.268857479095459, "identity": 0.2546731233596802, "hypothesis": "De de deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/560f5ab291f719dd00239232a6a745365c87f956e59223745602b7433129475c.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": "56760fe2d7110744aeed7c1a7f5f1cc0a2fa8902044b1f0b102dc897af9107f9", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "431a3b3f115df8a72267f9263c21e2d1c44ce6c58354fffd3bafed0e19f79518", "generation_seconds_batch": 16.352822346962057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4834939241409302, "target_z": 1.3828124319033661, "ce": 6.054384231567383, "genuineness": 4.362998008728027, "identity": 0.7612379193305969, "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_VOLT_reg_low/56760fe2d7110744aeed7c1a7f5f1cc0a2fa8902044b1f0b102dc897af9107f9.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "5695a4ebfbe3265a400d6fb632f91291c09457cf5b59196518aa1828e14df7c9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a0af5125c793bd026dbcf9fc51fa0cfc6d9eee8a60f7d2fb951429ebbd8ef00a", "generation_seconds_batch": 21.62253045500256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.062326908111572, "genuineness": 4.3127899169921875, "identity": 0.5938335061073303, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5695a4ebfbe3265a400d6fb632f91291c09457cf5b59196518aa1828e14df7c9.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": "56b1d87377627e245f4d9c4fa037ca1d7726c16ba5a2150921a2595b8ed3d1fc", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "788ef97bd4752244fb3ca8001f29f1153757c4dcd21c7f4bf7b3de4cb0f354b4", "generation_seconds_batch": 15.437324968981557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.23706579208374, "genuineness": 4.293961524963379, "identity": 0.723770022392273, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/56b1d87377627e245f4d9c4fa037ca1d7726c16ba5a2150921a2595b8ed3d1fc.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "56cef110ba1c24b72024c6c86caa1456deb4c330a8a9af2fce7acbf8ea4bf09a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f27c207b0693d90555b9dc53a7fd1722f8a88af1e1bc2c6b82a9f55509fbb3cf", "generation_seconds_batch": 17.66756415704731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.5443260669708252, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/56cef110ba1c24b72024c6c86caa1456deb4c330a8a9af2fce7acbf8ea4bf09a.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "56eda52231b2eaa2bce55a7287d2c13c5c3d784c79e4ffd5e623ebe69d4aaebb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc9683d32b58854d70aa908ce071a4d238b31fd4358200aff65b28734d43bf33", "generation_seconds_batch": 25.976079149986617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 5.879645347595215, "genuineness": 4.2249250411987305, "identity": 0.6210729479789734, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/56eda52231b2eaa2bce55a7287d2c13c5c3d784c79e4ffd5e623ebe69d4aaebb.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "572cd092cfc7b8278a5d4e8664066bf9174ee5f0acf71e4ff4a5ea912ca42e99", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0474c8eab011546673bf780cb801c1eba325f121ffeb864f4e77ab19b8659afd", "generation_seconds_batch": 16.40866323490627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3885064125061035, "target_z": 1.4921874908847026, "ce": 6.070269584655762, "genuineness": 4.507347106933594, "identity": 0.5621078610420227, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/572cd092cfc7b8278a5d4e8664066bf9174ee5f0acf71e4ff4a5ea912ca42e99.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "574098003ac589c93baff20cd391a52e19e7577dd5ed9a3188121e0573085381", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5911c2aa2a5a89e942eb10a9437d3493fd10775d308e8f56623ac33d280ecaf5", "generation_seconds_batch": 21.04820782900788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.205295085906982, "genuineness": 3.8671910762786865, "identity": 0.2895642817020416, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/574098003ac589c93baff20cd391a52e19e7577dd5ed9a3188121e0573085381.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": "590ba37ece575739b7705ce7a9590f5279730db961e59c0b5419dc7218f06a85", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Escribí el número en el sobre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6bd56028132ea3fc5243d14649b7bb3efcc55812b402bcb6f15644da1a89112", "generation_seconds_batch": 22.647255984018557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.125868320465088, "genuineness": 3.8922953605651855, "identity": 0.6947538256645203, "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_VOLT_reg_low/590ba37ece575739b7705ce7a9590f5279730db961e59c0b5419dc7218f06a85.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "5992812214eb338eca8e2b926219e34efc61e3883bedabf4e00dbd3a44ad435d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "e65cedd96d2e60f0591b2edb56564f56fb599ac4859ba8c4a5fc628e17123387", "generation_seconds_batch": 22.269368395907804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 6.141753673553467, "genuineness": 4.6203155517578125, "identity": 0.7888199687004089, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle connaît.", "wer": 0.05263157894736842, "duration_error_s": 0.120000000000001, "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_VOLT_reg_low/5992812214eb338eca8e2b926219e34efc61e3883bedabf4e00dbd3a44ad435d.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "5a3ecf51e402faeade7fef4aa7b76a75e6c54b8e39c65e0941d6fc0a29eb91f4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9105196c0b543bce88074ce761ebe45b31bd41337b16a2eb9a476eef2ac761eb", "generation_seconds_batch": 22.546399872051552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.435632705688477, "genuineness": 4.319066047668457, "identity": 0.5934476852416992, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5a3ecf51e402faeade7fef4aa7b76a75e6c54b8e39c65e0941d6fc0a29eb91f4.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "5a7ce5aae054c2471f470d8920e46812a96d56f28ff434542674b66f6e850afb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4f41d0437db81da6e19f696d38df423396e5ebfe1c73bfb86cbb13ae1e7274b", "generation_seconds_batch": 17.075972283957526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.103543996810913, "target_z": 1.820312530563056, "ce": 6.252951145172119, "genuineness": 3.9048471450805664, "identity": 0.3327583074569702, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5a7ce5aae054c2471f470d8920e46812a96d56f28ff434542674b66f6e850afb.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": "5aca90c9d1ab71e1a6cfad83b0f6da9b0e750dc2df9e844c62be14a84b30334b", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f9dcc6a832c871b28f28412e5d325b83c45cb067c4a0e1b14c85572ed666433", "generation_seconds_batch": 22.346272171009332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 5.967014789581299, "genuineness": 4.432034492492676, "identity": 0.8150664567947388, "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_VOLT_reg_low/5aca90c9d1ab71e1a6cfad83b0f6da9b0e750dc2df9e844c62be14a84b30334b.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "5ad6eee8bda17176504f3b1c6ab4ab30dadd29689c648b45762e0493aab2bc3f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "12ae66d122aa7fcca818342abcb1267ffced5d2a7fe7dd63aaa9a9db5f7b9d52", "generation_seconds_batch": 19.633100384962745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.3641486167907715, "genuineness": 4.168440818786621, "identity": 0.6207727193832397, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschneben.", "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_VOLT_reg_low/5ad6eee8bda17176504f3b1c6ab4ab30dadd29689c648b45762e0493aab2bc3f.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "5b51fad740d8b06ea9e19d77df7629db753d2f5bb161448dba49ad4276157ec6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "febd99b2a27e04dfd70f50fd305912358fce4bf045fe2c4ea1cea1516fcb3267", "generation_seconds_batch": 15.755795014905743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3070886135101318, "target_z": 1.5859374630026162, "ce": 6.213237762451172, "genuineness": 4.306513786315918, "identity": 0.3916206359863281, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5b51fad740d8b06ea9e19d77df7629db753d2f5bb161448dba49ad4276157ec6.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "5b8ffcdf531cf7da177ba722e12bf3bde2187f2dcef0e8aa4484c49afb8b3ad1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e386d8e82aa73e8e60e7de0a37b98d9decab1318efba52261ca01844552a4541", "generation_seconds_batch": 17.022553784074262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.062326908111572, "genuineness": 4.080576419830322, "identity": 0.6675442457199097, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5b8ffcdf531cf7da177ba722e12bf3bde2187f2dcef0e8aa4484c49afb8b3ad1.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "5bbe4998c1147eb4c96daed185da0ee3b64eadbcd6361b0a6d2fb4568be5c1e1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c764708838f263167a799b88ed5cbb79f66e8c4a9933697ca1393567d721aa5", "generation_seconds_batch": 15.606603272026405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.403861999511719, "genuineness": 4.068024158477783, "identity": 0.6616889238357544, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5bbe4998c1147eb4c96daed185da0ee3b64eadbcd6361b0a6d2fb4568be5c1e1.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "5c0f3bca64631865083bae66efe7582fcecc2e73e3f876ba75d854bea98dca78", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab88134d4fcca17a1f4fc1d2b36f95b22aa6291cfbbbc3c020ac3a66b434b298", "generation_seconds_batch": 15.44864875706844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.356205940246582, "genuineness": 4.13078498840332, "identity": 0.4424082636833191, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5c0f3bca64631865083bae66efe7582fcecc2e73e3f876ba75d854bea98dca78.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "5ca1edc7da49bbea1bc0e46214edeee13bade2ad167492bddd8dd8516fbfd659", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c426410526069e89aa92b1b6356c99118b3fac011511d5a926f165464b2d049c", "generation_seconds_batch": 24.889232890098356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.7747082710266113, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5ca1edc7da49bbea1bc0e46214edeee13bade2ad167492bddd8dd8516fbfd659.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "5cfe0fde57d3482e5e23f463eff5d0cc4ba9ff7da7c7ddc913126dabedb6891e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdde87ac9a000e8268de27b9e3e2be9772d7dc6cd1a0ffe21e5878302e31c769", "generation_seconds_batch": 15.055178398964927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.252951145172119, "genuineness": 3.9676074981689453, "identity": 0.3937165141105652, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5cfe0fde57d3482e5e23f463eff5d0cc4ba9ff7da7c7ddc913126dabedb6891e.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "5d5fe8035d884fadc7c4fa17ae771cc7299db9ca8c612018476141cb161dd540", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbc0ef00281e3e41f2af8240f36bde915cbca5fb2e4afa97bab5b3e3d39141c0", "generation_seconds_batch": 19.972900951979682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.403861999511719, "genuineness": 4.319066047668457, "identity": 0.39140668511390686, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5d5fe8035d884fadc7c4fa17ae771cc7299db9ca8c612018476141cb161dd540.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "5dedd4d578c85394c018aa8d4cfc15fcc522688f75d660d7d31566ebed1ebc23", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61c05bc1f7bd398601e0dc438d3f20826e622e85d475947de40713a9a628272e", "generation_seconds_batch": 15.710697405971587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.3765438199043274, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5dedd4d578c85394c018aa8d4cfc15fcc522688f75d660d7d31566ebed1ebc23.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "7c15a06b0903eade5d139423d168019f738b8d50f547efbeea394349cab97f76", "generation_seconds_batch": 13.594493446056731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.213237762451172, "genuineness": 4.557555198669434, "identity": 0.770767331123352, "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_VOLT_reg_low/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "5e8c8748021a8f5a01de76a4ce4fc70cdcb4ea3dda7530237a9533036ad3ea05", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "687684c3ded017d1bfed7009a708af8d693b3b9688ceca5f9bb47ae5364e9869", "generation_seconds_batch": 17.022553784074262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.768456220626831, "target_z": 1.0546875294906681, "ce": 6.229123115539551, "genuineness": 4.337893962860107, "identity": 0.36193785071372986, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5e8c8748021a8f5a01de76a4ce4fc70cdcb4ea3dda7530237a9533036ad3ea05.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "5eb822c33e09d3eca1a0b23bedd638dfc107d1f94d9a3f187f934c88e5a760e7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33d07ee60a6776d10c1b83f652326b8b6a85fe95195f5575b86db25e26700202", "generation_seconds_batch": 21.97986809106078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8253664970397949, "target_z": 2.1406249581768706, "ce": 5.839932441711426, "genuineness": 4.293961524963379, "identity": 0.768944263458252, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5eb822c33e09d3eca1a0b23bedd638dfc107d1f94d9a3f187f934c88e5a760e7.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "5eea54a06147b4970a1fbfb227204073551df3e48e0447fedf05f22ef17facc6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccdf3e29f1bc86dbf521cbeb63d53421127d622fabfa6dd74ec8a53e0f6001a1", "generation_seconds_batch": 23.386858014040627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.266379714012146, "target_z": 1.632812449061573, "ce": 6.109982967376709, "genuineness": 4.080576419830322, "identity": 0.7242075800895691, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5eea54a06147b4970a1fbfb227204073551df3e48e0447fedf05f22ef17facc6.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "5f4c64efe37d61f40abda3dfafc84212f1d9ddce7ddd1e5472e5bb1760abfc16", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "449cd6676b367477f791147c00a9c20d63dd43aa456b04f70552bbd6ce8033c0", "generation_seconds_batch": 20.303445564000867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.1020402908325195, "genuineness": 4.20609712600708, "identity": 0.6321725249290466, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5f4c64efe37d61f40abda3dfafc84212f1d9ddce7ddd1e5472e5bb1760abfc16.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "5ff4f5071a0b65309f4098a296f16da3cf0dfc07a2bd197a4e21d39cebbca168", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "711e4cbc37b0761cc491099ac164b511a5f7a14d44914bb2d52ca924edc7644a", "generation_seconds_batch": 22.67651059210766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.141753673553467, "genuineness": 3.798154592514038, "identity": 0.540228009223938, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/5ff4f5071a0b65309f4098a296f16da3cf0dfc07a2bd197a4e21d39cebbca168.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "610d22c1902037215edfe6a9052322fb8a95337e517b3fdf596ac15901451e31", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ca7ac4624ba0c2e21c853b4a865a3bcb9191b52ade15ad7d272052e75efebbc", "generation_seconds_batch": 16.705525048077106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.3641486167907715, "genuineness": 4.036643981933594, "identity": 0.6856917142868042, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/610d22c1902037215edfe6a9052322fb8a95337e517b3fdf596ac15901451e31.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "6238548d0cc0392d48ac488a33d46ac1664b29ba67bf14168907d4b11f87926e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16b7d8962e2ace316ba71d357a8c76625d79f6fa1a471914e7f563518509b95e", "generation_seconds_batch": 16.691507424926385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.23706579208374, "genuineness": 4.469690799713135, "identity": 0.7610825300216675, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6238548d0cc0392d48ac488a33d46ac1664b29ba67bf14168907d4b11f87926e.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "623d0adbb62de2b885917d0adea6ef7d34028cd2900d6020ef00d78bc35ccdb6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ebe0bb478e214284cb6886def8e3ea0dab84d45ac96cd4b0d9d1661407aed05", "generation_seconds_batch": 15.801718350034207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2935189008712769, "target_z": 1.601562549866039, "ce": 6.0305562019348145, "genuineness": 4.394378662109375, "identity": 0.6215012073516846, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/623d0adbb62de2b885917d0adea6ef7d34028cd2900d6020ef00d78bc35ccdb6.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "62759f5f6399882e42b6563bf08cbbadf9012c09deb185c4368cdc43e0d30f80", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "460e9c288aee8944050bbc0fcaa81af522140506fdce9bc8a582e9b8d9495d3a", "generation_seconds_batch": 16.377679045079276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5784813165664673, "target_z": 1.2734375101876854, "ce": 6.086154937744141, "genuineness": 4.570107460021973, "identity": 0.5934592485427856, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/62759f5f6399882e42b6563bf08cbbadf9012c09deb185c4368cdc43e0d30f80.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "62bbec613bbf9f30a4cb4fc0b04a677f7092b7f2f2daa02d93001038c30fc499", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7de4bbc8839515f69950129a2fe1e09da62f93b1e88f8708f185d6050d0cbc78", "generation_seconds_batch": 15.24645802390296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5445572137832642, "target_z": 1.3125000214477587, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.6187304258346558, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/62bbec613bbf9f30a4cb4fc0b04a677f7092b7f2f2daa02d93001038c30fc499.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": "62bdf71acef6aebbddf456fc23d278dddb4223290bb8d1c954ed8d51b366035c", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb014835693a0c52c424470f7cf1100027a82c12ab2c1f55d22c71b13841cf7b", "generation_seconds_batch": 16.085609073983505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.1894097328186035, "genuineness": 4.281409740447998, "identity": 0.8108398914337158, "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_VOLT_reg_low/62bdf71acef6aebbddf456fc23d278dddb4223290bb8d1c954ed8d51b366035c.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "638d44968373eaf5ae02debe88fb7a1b721b07b30863a48807e9627893146a1a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1de4efbbe0e35e712f7d7e054a502816146603951d6510e06735d6d0e8b827c0", "generation_seconds_batch": 21.97868969803676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2799493074417114, "target_z": 1.617187499463806, "ce": 6.24500846862793, "genuineness": 4.005263805389404, "identity": 0.6845070123672485, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/638d44968373eaf5ae02debe88fb7a1b721b07b30863a48807e9627893146a1a.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "63bf2ac7b880029fb62ee28d20df3456c35946ca502311d4d5b891bd94710288", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45220d9791d0cbc2bfe2388f98fa9fec6670fbf14af8681769985484fad9926e", "generation_seconds_batch": 23.912634007050656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7032396793365479, "target_z": 2.281250053619397, "ce": 6.284721851348877, "genuineness": 4.400654315948486, "identity": 0.7411891222000122, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/63bf2ac7b880029fb62ee28d20df3456c35946ca502311d4d5b891bd94710288.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "6452d70b01aa520f18f9ca69b9b8d4d0f6d7af2ef7490cc5c9427f57aaf25ebb", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52aa8a7b992a1dbbe45da46263ee8994b8903f4c7eb3431a766a47cf52cc2220", "generation_seconds_batch": 15.930861535016447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.125868320465088, "genuineness": 4.218649387359619, "identity": 0.6694393754005432, "hypothesis": "ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6452d70b01aa520f18f9ca69b9b8d4d0f6d7af2ef7490cc5c9427f57aaf25ebb.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "64e2a6a5c00b2ca41d4daf92e0bed55b1cb2938de5f60cd8a2538cf9ce62a4d7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1605e5a8a77084cd2eaa81d8d21ae66a791b41e412d959004684bb8f2f5f694c", "generation_seconds_batch": 21.393748603994027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088608026504517, "target_z": 1.468750066488052, "ce": 6.38003396987915, "genuineness": 4.3881025314331055, "identity": 0.6731500625610352, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/64e2a6a5c00b2ca41d4daf92e0bed55b1cb2938de5f60cd8a2538cf9ce62a4d7.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "64f942d194a011349158385b7bb37fa8f69b07bc5e3a38f7314564c070106df5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2e7719c05a9a808475ef56bcd9e391ef5a37d37a8156db22f7f64c7e36c7e3ac", "generation_seconds_batch": 16.04618112789467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.1894097328186035, "genuineness": 4.331617832183838, "identity": 0.5596979856491089, "hypothesis": "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_VOLT_reg_low/64f942d194a011349158385b7bb37fa8f69b07bc5e3a38f7314564c070106df5.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "65582ef1f6736c0f6239365cd3ce022de78cf18b90cda45654c5954791b153b1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "016d556601bbe8c4d1f49eaaf79c50f746b3e9678fb227413740f7d032ac5740", "generation_seconds_batch": 16.321099309949204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.157822608947754, "target_z": 1.7578124576406766, "ce": 6.09409761428833, "genuineness": 4.093128204345703, "identity": 0.5606273412704468, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/65582ef1f6736c0f6239365cd3ce022de78cf18b90cda45654c5954791b153b1.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "659be8b2d0fb4959ba3f963e51821674fe868f0d9ce89824d2ba9cbc53723a92", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1392f245657b8cd59eae69d9d842894851f0ced80441618dcd6b4b1f37282cf9", "generation_seconds_batch": 15.628013657988049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.395919322967529, "genuineness": 4.250029563903809, "identity": 0.491152822971344, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/659be8b2d0fb4959ba3f963e51821674fe868f0d9ce89824d2ba9cbc53723a92.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "66b9057590d5b88dbed706d153f193818125e9deb0d6d11e54f3d97912086bec", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed10fea651e7a5faec2394840adba817f10c4e480c7c4f3cde041c7502bdf9e1", "generation_seconds_batch": 21.251775570912287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5038483142852783, "target_z": 1.3593750075067155, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.21021829545497894, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/66b9057590d5b88dbed706d153f193818125e9deb0d6d11e54f3d97912086bec.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "66fe8e3367d8429935b5c9b15c379756dadc5a4280afbdcbc1681aaac2254595", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ba5fa566aa2019fb3846602ea273870e9dcd1588726f78d6686d6b330befb83", "generation_seconds_batch": 17.65645062807016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.213237762451172, "genuineness": 4.017816066741943, "identity": 0.2641620337963104, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/66fe8e3367d8429935b5c9b15c379756dadc5a4280afbdcbc1681aaac2254595.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "671958fabaeb169442a13cd37cbaff267518354222f2f26ca7742f75b75c326d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ae42e3571aed1901f00ff6ca70701b8680cb777809e887769891fde170b882b", "generation_seconds_batch": 20.316671199980192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.3641486167907715, "genuineness": 4.256305694580078, "identity": 0.4083109200000763, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/671958fabaeb169442a13cd37cbaff267518354222f2f26ca7742f75b75c326d.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "67e32c99955a0a1a54378b6dc32d9ab62bd6508ebe14c1127f90d553cd8b572e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "487555e505b873ac522a52351c36051d2ecaa1d5167b556a4aec4024deb4717c", "generation_seconds_batch": 23.088737585931085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.1894097328186035, "genuineness": 4.362998008728027, "identity": 0.6936537623405457, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/67e32c99955a0a1a54378b6dc32d9ab62bd6508ebe14c1127f90d553cd8b572e.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": "6813725134bddb6381b1b71baa7eeedc17331e263a4cfb579546393557e1bf41", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f647e52c25d5523779d177287f4fa90106ef18342dfdc7405345178eaad38c60", "generation_seconds_batch": 16.388820045045577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.165581703186035, "genuineness": 4.607763767242432, "identity": 0.6561475992202759, "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_VOLT_reg_low/6813725134bddb6381b1b71baa7eeedc17331e263a4cfb579546393557e1bf41.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "681fab72225da3107fdb2bcb5f450f1dabf4fff03f48ffbe7f43aa7a0d09e591", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b439b91f9eaf42b4974f0ce9597036516700b5e3e75f108bb94bee0e0909e58", "generation_seconds_batch": 22.961514586000703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.070269584655762, "genuineness": 4.519899368286133, "identity": 0.5616904497146606, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/681fab72225da3107fdb2bcb5f450f1dabf4fff03f48ffbe7f43aa7a0d09e591.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "682596182f32f724fe015cbb2efd906dc47fe0bf35ed74b9276a47a70011cbf9", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d19217c9aad61e69e97697dde139728e431d071dedea5e9a38e4e519361a9239", "generation_seconds_batch": 17.199576804996468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.403861999511719, "genuineness": 4.1998209953308105, "identity": 0.5331898927688599, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/682596182f32f724fe015cbb2efd906dc47fe0bf35ed74b9276a47a70011cbf9.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "686c7492744250657872b7a8c551008952ee692efbd7c403c401c6def1dfde4b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c52fde5525b7ab4f293e36823d190fd2cd774544a2a252d900b30f42fc2e747", "generation_seconds_batch": 21.585452035069466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.078212261199951, "genuineness": 3.9801597595214844, "identity": 0.5097814202308655, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/686c7492744250657872b7a8c551008952ee692efbd7c403c401c6def1dfde4b.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "68836ccba77d3d15b7bf139376973a98a307b7cf5dd2cf23fb2d63aaa62ec6c7", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "27fd863043171a5244c00e51cc1c210fb43bf403a4d8be7879be127037b2af3c", "generation_seconds_batch": 15.90894788398873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.046441555023193, "genuineness": 4.093128204345703, "identity": 0.33403897285461426, "hypothesis": "Ik 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_VOLT_reg_low/68836ccba77d3d15b7bf139376973a98a307b7cf5dd2cf23fb2d63aaa62ec6c7.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "68d8a740fa052b5c5fd1393964a980080eaaf76f04aba6d54b01b93f86bfc6dc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "497031439f0854acc505152db27babe20e3ca44d1de2d6494ca4e0c067b9bc12", "generation_seconds_batch": 22.971873180940747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.221180438995361, "genuineness": 4.369274139404297, "identity": 0.5031520128250122, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/68d8a740fa052b5c5fd1393964a980080eaaf76f04aba6d54b01b93f86bfc6dc.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": "68f38954792c32cc8b355a20d74dbbbc879fb756b285ed2ad192ea67b4b3adbe", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a635c069c4ad56b99cd87781c35ad0d4ada8f2e6899fd07c7420c81335579edf", "generation_seconds_batch": 16.388820045045577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.24500846862793, "genuineness": 4.545003414154053, "identity": 0.7616077661514282, "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_VOLT_reg_low/68f38954792c32cc8b355a20d74dbbbc879fb756b285ed2ad192ea67b4b3adbe.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "68f5d5df2af0437a9b724e7d73c6b2825f9e8bd6855e8b47d389ecfe27b0e760", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf13a1323042d90d77850ddc1474bbce3c25912518c290efa645af50895d8b7d", "generation_seconds_batch": 14.818628950975835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.7958021759986877, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/68f5d5df2af0437a9b724e7d73c6b2825f9e8bd6855e8b47d389ecfe27b0e760.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "691dc1b7b495b55c90667f3ed04624876c8086014ef87cbbfe225ac4708b0849", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "652f0b5cad1e1626258bf72f209df4bbc985b4478ab6f6ee7c4439c03eadb6d3", "generation_seconds_batch": 22.37410696398001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3342278003692627, "target_z": 1.5546875638070823, "ce": 6.427689552307129, "genuineness": 4.055472373962402, "identity": 0.6232151985168457, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/691dc1b7b495b55c90667f3ed04624876c8086014ef87cbbfe225ac4708b0849.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "69398b48b9236351e2dbf05cbae284f63feb87e8ee8a07622b9029dc02367d34", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ecf45f44b40090a785bb9dd834b25b68c78a98ad4c3bfe066a77a0df9a05948", "generation_seconds_batch": 16.946781497914344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.6633187532424927, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/69398b48b9236351e2dbf05cbae284f63feb87e8ee8a07622b9029dc02367d34.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "6983d5f4093b543576ecc7bb3336eb473356419e30a85dfe141693707e10db9d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db3d8ad668c379ba022340edd8e0987361e3c0dcd489ac6cb591cac27abd92c8", "generation_seconds_batch": 16.49261487601325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.8059420585632324, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6983d5f4093b543576ecc7bb3336eb473356419e30a85dfe141693707e10db9d.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "69c556af8d1b70d9506844643513795c2613046a6ac6e8758de70a0bf1f37f78", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1e0e4fd42f302248571b8eb31d987fd2f6e48166fabf8689c08975f77a3bb5f", "generation_seconds_batch": 21.474002231960185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 5.696964263916016, "genuineness": 3.9048471450805664, "identity": 0.46849143505096436, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/69c556af8d1b70d9506844643513795c2613046a6ac6e8758de70a0bf1f37f78.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "6a2550dd9fcd9064ef935c56f09b0d7546a330ff34ef4d35d11959b8b8af75f6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48ae9dda2c346cdf2490f7533c66b3dbc96e1cf1167a41362c16730121712882", "generation_seconds_batch": 22.038292304961942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5920509099960327, "target_z": 1.2578125605899184, "ce": 6.062326908111572, "genuineness": 4.231201171875, "identity": 0.6795427799224854, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6a2550dd9fcd9064ef935c56f09b0d7546a330ff34ef4d35d11959b8b8af75f6.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "7ff90b0b2d600a8cdec6152961e06c02042df66cca670443fc5a49c3f3ba065b", "generation_seconds_batch": 16.014489284018055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.316492557525635, "genuineness": 3.9927117824554443, "identity": 0.24742676317691803, "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_VOLT_reg_low/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "6a69c9cb45292c97afdfb9d2925b0f538f5758792b7ac0a54e70d4bced388d16", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac0cefa7058f552e7a6bf9b33327726045b882a78cb5da0c8011eb09131cd888", "generation_seconds_batch": 20.98010233498644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3070886135101318, "target_z": 1.5859374630026162, "ce": 6.221180438995361, "genuineness": 4.2374773025512695, "identity": 0.42595526576042175, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6a69c9cb45292c97afdfb9d2925b0f538f5758792b7ac0a54e70d4bced388d16.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "6a9356d970df1f0fda9d26f4a29e282dae6e2a8fb0d6ae659a08b6f9a282640a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcb7bcbc147da7e7f4106ca43c3b61d4c1a846499de5c8ea8df3348341e7c75c", "generation_seconds_batch": 21.852399035939015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.6566474437713623, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6a9356d970df1f0fda9d26f4a29e282dae6e2a8fb0d6ae659a08b6f9a282640a.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "6ad0ff8bdba830dce858fbe2c75fc591db039176df5e9f38aead330423157d09", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1615c47f5b70290b42a1d1f618ef86d83ecc0e596e43422a568e079af50c364b", "generation_seconds_batch": 16.691742539056577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7168093919754028, "target_z": 2.265624966755974, "ce": 6.0305562019348145, "genuineness": 4.3881025314331055, "identity": 0.7609995603561401, "hypothesis": "La porte 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_VOLT_reg_low/6ad0ff8bdba830dce858fbe2c75fc591db039176df5e9f38aead330423157d09.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "6afed1cc3500cab8e26fb48b4d9ad009a245f6ef7b268f48e53fab2233fe9710", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a94c42bf180be6e0599fe2abd344973b5cfb6b9a7d4774477a54ceb0b7df5a0a", "generation_seconds_batch": 17.18187458603643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.308549880981445, "genuineness": 4.394378662109375, "identity": 0.39101535081863403, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6afed1cc3500cab8e26fb48b4d9ad009a245f6ef7b268f48e53fab2233fe9710.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "6b4502cab002999a3baf51408098f490149666efb9c800412eeea64c0b4af438", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0529930edb86ae0394178a890707aba7c48e678e7cc515f55880e4c49fa7912", "generation_seconds_batch": 16.962318576988764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.268836498260498, "genuineness": 3.898571252822876, "identity": 0.3930066227912903, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6b4502cab002999a3baf51408098f490149666efb9c800412eeea64c0b4af438.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "6b6a14703f187eba983a591ef3d4d3270f119bf895bd71ba9d2a2b44a9765b45", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b35d6b6e1cfa6787ae3e4707033326ec4738e6c8efc6a7024727640d76a5de9d", "generation_seconds_batch": 23.715554426889867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.395919322967529, "genuineness": 4.093128204345703, "identity": 0.42543256282806396, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6b6a14703f187eba983a591ef3d4d3270f119bf895bd71ba9d2a2b44a9765b45.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d3c8205c1ed893fb856ff1f9d929300a8ff7aacb6d1bedcf879ae5d638d5c375", "generation_seconds_batch": 18.816525902948342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5174179077148438, "target_z": 1.3437500579089485, "ce": 6.141753673553467, "genuineness": 4.7332844734191895, "identity": 0.6268541216850281, "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_VOLT_reg_low/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "6bdaeae442efca23c3c6a31eb5d1da66e59fea754543537e1264bc0f2862e860", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cc813f59797166b1e2215e8c510ff7e852515afca1891c5825c271155369043", "generation_seconds_batch": 22.90327510307543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.078212261199951, "genuineness": 4.375550270080566, "identity": 0.7171652317047119, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6bdaeae442efca23c3c6a31eb5d1da66e59fea754543537e1264bc0f2862e860.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "6c1cc7d53956057aa4f278822811a23e87b0013122c5ca96c1faf54a2a07dc0f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0c2248f1dcbb83d7fea711542fa8eae16d462e029fd750f85cedcf7deb48de35", "generation_seconds_batch": 15.755795014905743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.443574905395508, "genuineness": 4.162164688110352, "identity": 0.35596442222595215, "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_VOLT_reg_low/6c1cc7d53956057aa4f278822811a23e87b0013122c5ca96c1faf54a2a07dc0f.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": "6c8d1cb1159440aca706b72e34f7743340d2b66a066c9c97214892bdbd0a3768", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7153c8c568338035cd075b14f851dab80a8abf8348013c4b33cf5856f5a1eb6a", "generation_seconds_batch": 19.30227281700354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.0305562019348145, "genuineness": 4.3504462242126465, "identity": 0.6209255456924438, "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_VOLT_reg_low/6c8d1cb1159440aca706b72e34f7743340d2b66a066c9c97214892bdbd0a3768.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "6cbfdbec2fa3a409a87f422b03cdac158f7c15c6a47f27203c8c45401ab38239", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da5f7ad6f57a4b7199a6f1e63018b97360829726a316974d15b6fbfcc7be88a2", "generation_seconds_batch": 21.65375940897502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 5.895530700683594, "genuineness": 4.356721878051758, "identity": 0.677626371383667, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6cbfdbec2fa3a409a87f422b03cdac158f7c15c6a47f27203c8c45401ab38239.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": "6cc3a8a39daca47afb0c5017ef6b5422fbbdc0f366ad3b4ccbe9ca13167218d9", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1784c50d4fc71cf23cdd3de791d70a3631a1b5b7d237344308de167274042165", "generation_seconds_batch": 19.039436175953597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.308549880981445, "genuineness": 3.961331605911255, "identity": 0.35033515095710754, "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_VOLT_reg_low/6cc3a8a39daca47afb0c5017ef6b5422fbbdc0f366ad3b4ccbe9ca13167218d9.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "6cf98f89bc6d7a7d4c32ffe2f848d5989d63152d521f9efaf2c4e75328c15b8c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d8d574778a7cde00550d68bfa9326c3d1870634ded1caf12b4faef10a54e795c", "generation_seconds_batch": 21.410446002963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.653226912021637, "hypothesis": "J'ai trouvé la 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_VOLT_reg_low/6cf98f89bc6d7a7d4c32ffe2f848d5989d63152d521f9efaf2c4e75328c15b8c.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "6d83a28f5e8324f4699257d76bf2ee70d214dd35e82cf3e70aa0ee3acb88b67b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d530afd778a77a6567b1957fc91298ccc238adf510dbc3dbd691d4dab5e2317", "generation_seconds_batch": 16.818142972071655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6259750127792358, "target_z": 1.218750049329845, "ce": 6.356205940246582, "genuineness": 4.494794845581055, "identity": 0.4413435459136963, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6d83a28f5e8324f4699257d76bf2ee70d214dd35e82cf3e70aa0ee3acb88b67b.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "6de196c7525bf516c0caf9c7873654f2c9e15ff1ce301ae61bd61456db4ae238", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65e8fa11a3d18e9ce5abe88e65cf19608dd323936e4fbb7b2c4812bd8536683d", "generation_seconds_batch": 20.024281559046358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.197352409362793, "genuineness": 4.344170093536377, "identity": 0.5615476369857788, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6de196c7525bf516c0caf9c7873654f2c9e15ff1ce301ae61bd61456db4ae238.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "6df2d34aecb5baa53b0e471b7967552913700196815836c0b42deda86b3a25df", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9bf123fc0df0279128264365e59ca0ba9119172a9e54d8662be78fb62c9caf6", "generation_seconds_batch": 14.994901362922974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.2767791748046875, "genuineness": 4.168440818786621, "identity": 0.6402897834777832, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6df2d34aecb5baa53b0e471b7967552913700196815836c0b42deda86b3a25df.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "6e071291c212da235172ebb8ba35f4ecd90313f81c6e9a3dff5e5263da4728e5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb0764f1d57a82258ab3c558c59c04d5985e0a5b593bee5d4df2ced8ea4bcc2b", "generation_seconds_batch": 20.303445564000867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3885064125061035, "target_z": 1.4921874908847026, "ce": 6.0146708488464355, "genuineness": 4.457139015197754, "identity": 0.6827899217605591, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6e071291c212da235172ebb8ba35f4ecd90313f81c6e9a3dff5e5263da4728e5.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "6e19ac9284fd107430dbe6e81661bc647c6ca00e5ae83a0a7a110d57cecfbe25", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "25bd48ba6562685b3b3d0e42a63974b22c708282e212d2604fb2d2393aae8fcb", "generation_seconds_batch": 22.79530879703816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 5.9431867599487305, "genuineness": 4.3881025314331055, "identity": 0.7350192070007324, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6e19ac9284fd107430dbe6e81661bc647c6ca00e5ae83a0a7a110d57cecfbe25.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "6ead1e7522e5292ae44e1b0a5ec35f08860ef49a406f0f2ae87d5cb2af57115b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17f66f32d99237a5c7396cac532cce3280274b499d9d0c41bef194a0960baaee", "generation_seconds_batch": 20.316671199980192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.3641486167907715, "genuineness": 4.268857479095459, "identity": 0.40524381399154663, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6ead1e7522e5292ae44e1b0a5ec35f08860ef49a406f0f2ae87d5cb2af57115b.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "6f03c530fb7d903d16d7317fdf3d382f6c3000630390de97ea61831f4eab8baa", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c590f1ac81d8862c4a8094f84eafdec3899302c2cbd02e62c86303c41fc2340", "generation_seconds_batch": 18.937331393011846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.316492557525635, "genuineness": 4.532451152801514, "identity": 0.6164055466651917, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6f03c530fb7d903d16d7317fdf3d382f6c3000630390de97ea61831f4eab8baa.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "6f79a6b0e91afc83aeee90e86915acb1704766f0927167da1ba696050bf540a7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f1e49c5916ac901e5a8ad7abae19b1dbca32d82ed08d506ba4b19fc82687f77", "generation_seconds_batch": 21.210514620062895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.23706579208374, "genuineness": 3.9173994064331055, "identity": 0.25156670808792114, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6f79a6b0e91afc83aeee90e86915acb1704766f0927167da1ba696050bf540a7.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "6f8e58b41201864246998a64a2d65e669863c04495c1a72b6862b5d9522c727a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c4de642bbe38904e80a02519d92614339d1a9349eda36f38f92ed55b127633d", "generation_seconds_batch": 15.595161181059666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.197352409362793, "genuineness": 4.099404335021973, "identity": 0.7821453809738159, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6f8e58b41201864246998a64a2d65e669863c04495c1a72b6862b5d9522c727a.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "6f90a449f8fe570c2412b1064c5ee93d7ff64aa5ec62caef8311d265897100d2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1ee4bf140b53da8e2593ea75510e38ba660404862b1203702c6772435e64ce6", "generation_seconds_batch": 15.031020099995658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3817216157913208, "target_z": 1.499999965683586, "ce": 6.284721851348877, "genuineness": 4.400654315948486, "identity": 0.5937678217887878, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6f90a449f8fe570c2412b1064c5ee93d7ff64aa5ec62caef8311d265897100d2.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "6fa0bb0ae8a3cf0cb3a548b42d9e86cd9457081fc0a10797dd3521c2bcb0a70a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d49b97c0fe3d1096e277f932127f94e9f24f87179bc59fb3e2e63a4aa54c2d1", "generation_seconds_batch": 22.11684586200863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.38797664642334, "genuineness": 4.457139015197754, "identity": 0.5489310026168823, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6fa0bb0ae8a3cf0cb3a548b42d9e86cd9457081fc0a10797dd3521c2bcb0a70a.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "6fb9e9cb88abf3577a50e36943687ef7a79f0a9dcc15d077dd518e8f24953f87", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9490b7ce703666086611471e04b3692383a01485e28f2aa03758f2bd5a7f079d", "generation_seconds_batch": 22.479444041033275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.213237762451172, "genuineness": 4.231201171875, "identity": 0.7153196334838867, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/6fb9e9cb88abf3577a50e36943687ef7a79f0a9dcc15d077dd518e8f24953f87.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "7087b2002bcccae17f23c8d57031749c5a9059ea6d4b8f89b70a310187ead01a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "32454534941ff0f16d3a8ec18cb5b6edb578ee07235a91887e1cb65b3091571d", "generation_seconds_batch": 21.930645839078352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.372091293334961, "genuineness": 4.250029563903809, "identity": 0.6922836899757385, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7087b2002bcccae17f23c8d57031749c5a9059ea6d4b8f89b70a310187ead01a.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "70f24840266a7d913bf5b903c569aaed78d73d3f1cc5b37bed414461d49cdc4b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0184b1ed2a433f9edfc055fe6183a51c61a6c5f844e0c1ff1938d1003b3ca000", "generation_seconds_batch": 21.97986809106078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6218218803405762, "target_z": 2.3750000257373105, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.7101614475250244, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/70f24840266a7d913bf5b903c569aaed78d73d3f1cc5b37bed414461d49cdc4b.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "70ff61b6cf24a591500f950541ac7db19b04b557c183a4fe201e32977d4b94b5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "56086777560b797bb162c342ea0eed0bf3a20282a9bcd713746c8988078e6413", "generation_seconds_batch": 21.322221384965815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.356205940246582, "genuineness": 4.432034492492676, "identity": 0.45294684171676636, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/70ff61b6cf24a591500f950541ac7db19b04b557c183a4fe201e32977d4b94b5.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "71252e692dce2a00756ebdd1b9ee501d8d4ce6f03077422a99f44bcef4985ea1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf0c32ea7ccaf8a962bdda1126fd91242de02d28853c018d7e816b35b4d2c9c1", "generation_seconds_batch": 17.39317965891678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 5.95112943649292, "genuineness": 3.9425034523010254, "identity": 0.6472470164299011, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/71252e692dce2a00756ebdd1b9ee501d8d4ce6f03077422a99f44bcef4985ea1.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "715b323a55d17af28c63f9d01172032a5f46ed45c915cc31da93f63e5a53d535", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1eefabb80c09d53fd9c5e99bee03e7f201897ed37b34e317a8be1279eacec9d", "generation_seconds_batch": 16.277106542955153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7209625244140625, "target_z": 1.1093749903485086, "ce": 6.1020402908325195, "genuineness": 4.507347106933594, "identity": 0.4761424660682678, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/715b323a55d17af28c63f9d01172032a5f46ed45c915cc31da93f63e5a53d535.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "716d8e4fdc5094e6304b11dee2267212b0844c475eaacb4f998840309e7ddfe6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d702d92de64ffc6110f27d04cecc087479882d6c543f7658488981e95a1b8bf6", "generation_seconds_batch": 16.448964911978692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.356205940246582, "genuineness": 4.268857479095459, "identity": 0.43646401166915894, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/716d8e4fdc5094e6304b11dee2267212b0844c475eaacb4f998840309e7ddfe6.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "720b4385a6e90285111a3add1d7dfcd3f9471cdeb0d71f22841a633d78150b4a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88665fa4f17824f53776214158c0b1a747b32dd0867cd4f668b52d35cf96647d", "generation_seconds_batch": 21.68426930694841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.149696350097656, "genuineness": 4.394378662109375, "identity": 0.4953985810279846, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/720b4385a6e90285111a3add1d7dfcd3f9471cdeb0d71f22841a633d78150b4a.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "7367f8b77dc08a8e828c6374ac6f52c802866b02ca88a7db8c5f1cf66dc801f5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33460283f0fc9e22c50d51223eabc78a58c82f0f484fea63cc82b138993e90db", "generation_seconds_batch": 21.424190304009244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 5.788304805755615, "genuineness": 4.055472373962402, "identity": 0.6604762077331543, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7367f8b77dc08a8e828c6374ac6f52c802866b02ca88a7db8c5f1cf66dc801f5.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "737be1b6bf2c2223fe017c0d840558b7143a4e2992eac4ca43e7d191774ac01b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bd7d7bafc5673e4447df0f1ad3b7ce4d3b441a120f73665ba57990c725e2305", "generation_seconds_batch": 21.89769274101127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.475345611572266, "genuineness": 4.356721878051758, "identity": 0.6240023374557495, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/737be1b6bf2c2223fe017c0d840558b7143a4e2992eac4ca43e7d191774ac01b.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "73db3f8e58ba6343e820a15428f7c1b175fc80437967b0e3ad13e08a01d08a4e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6701098a4de6a37a6b84ce74b83e85227c996cb56167692c12ec606d9d83b00", "generation_seconds_batch": 15.603602790972218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6218218803405762, "target_z": 2.3750000257373105, "ce": 6.1894097328186035, "genuineness": 3.537698984146118, "identity": 0.6522105932235718, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/73db3f8e58ba6343e820a15428f7c1b175fc80437967b0e3ad13e08a01d08a4e.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "74b1c9de3ca64fd91ae96443d3734f17ed1cccff434173aaf1b228f8c0b5ab24", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21b780ef42f6c203447d7af52dc16de48392036c5f5650c6b6f5d00077e5e351", "generation_seconds_batch": 15.695639044977725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.23706579208374, "genuineness": 3.9173994064331055, "identity": 0.40660348534584045, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/74b1c9de3ca64fd91ae96443d3734f17ed1cccff434173aaf1b228f8c0b5ab24.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "74e5379c728c75479ca93572efaa96d1b8d61dee687173bb3b5b0dcd2d72c260", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a429555a611ac00a010ac0c8114cb9f833edbb926bcf1cbad90a3946221e007e", "generation_seconds_batch": 18.26572094100993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.490278720855713, "target_z": 1.3749999571044826, "ce": 6.292664527893066, "genuineness": 4.2374773025512695, "identity": 0.33476024866104126, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/74e5379c728c75479ca93572efaa96d1b8d61dee687173bb3b5b0dcd2d72c260.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "750d2bf3f2b650055e12a0fc8c65d5d7881726b2f3b6029b2f40419aeb6f5891", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f332548e8484f0775a0b931700a50c4a8b1667e5c1cf05ea7e85a019802db62d", "generation_seconds_batch": 21.852399035939015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 6.038498878479004, "genuineness": 4.394378662109375, "identity": 0.684914231300354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/750d2bf3f2b650055e12a0fc8c65d5d7881726b2f3b6029b2f40419aeb6f5891.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "758101ad71d08228af830dcdfde686ee414b93341e42991b67e904c902b03f55", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccbd1c0d6239a6a5848a906ef400a16f7b8763fbc3c7436a057ebe25b5c95440", "generation_seconds_batch": 23.142584107001312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.133810997009277, "genuineness": 4.519899368286133, "identity": 0.531782865524292, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/758101ad71d08228af830dcdfde686ee414b93341e42991b67e904c902b03f55.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "75ca64807093079ceabcd2c04f1403cde768084685017de667912690332f90e8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7baf511a652501febc708bac53bf98d89dee2fe86904a2d31b2d0a9f9e53d959", "generation_seconds_batch": 23.725946736056358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0764046907424927, "target_z": 1.851562567024246, "ce": 6.284721851348877, "genuineness": 4.093128204345703, "identity": 0.22664861381053925, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/75ca64807093079ceabcd2c04f1403cde768084685017de667912690332f90e8.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "75f263af7d6e90454042d87e73ec569eef10d8c68943e4edd7f41f3e895403eb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6c9bba82e7a49e264707ac66de97eafc0fa36b43055cffac285155c919754d7", "generation_seconds_batch": 15.25579786801245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.6441572308540344, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/75f263af7d6e90454042d87e73ec569eef10d8c68943e4edd7f41f3e895403eb.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "76064822214d723efd9d20416a48d5642c55c603bf47d88a6867eed03b623df3", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2e61e59fe31670c96b4c2f335ba2a82b5240b19bec9c62a7528f1df6cd939be", "generation_seconds_batch": 15.915383031009696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0764046907424927, "target_z": 1.851562567024246, "ce": 6.252951145172119, "genuineness": 4.469690799713135, "identity": 0.7493520379066467, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/76064822214d723efd9d20416a48d5642c55c603bf47d88a6867eed03b623df3.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "76a8c44e606aed7e64e435c5504ddc5427b13e24a112cefcfec11bb1aee6b853", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "494c58f397bb815eefb045ac2653c9f9b67e3eb6b8206f61b15bb9b1d27ab14c", "generation_seconds_batch": 23.142584107001312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.141753673553467, "genuineness": 4.293961524963379, "identity": 0.6827477216720581, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/76a8c44e606aed7e64e435c5504ddc5427b13e24a112cefcfec11bb1aee6b853.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "76ea9026751e239b3a66e39665a8a178b949c07535d522650e9006b593989553", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0dddef05cfbf943a5adf06aeb3d7acc112b9f405b4dc79d367de6aa644021ca3", "generation_seconds_batch": 24.4185386080062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.284721851348877, "genuineness": 4.061748504638672, "identity": 0.45444583892822266, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/76ea9026751e239b3a66e39665a8a178b949c07535d522650e9006b593989553.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": "771eb0f298d0e962e3e9dc7ee00c3926f715d41d3f4c43f41d6fa45d6f1e2b15", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0dd531ad4f1e2997fe60259be786b73eb0c830fc811bdedd10ae97f04290ed04", "generation_seconds_batch": 17.261192742967978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.8471037149429321, "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_VOLT_reg_low/771eb0f298d0e962e3e9dc7ee00c3926f715d41d3f4c43f41d6fa45d6f1e2b15.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "77428683b2e8871c1210034227de5e058740e87212deb1def55114831949934e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d74cdc77431596ee308b211dc323093f0f877039946c4f4b1f303e9db0712d8", "generation_seconds_batch": 21.21288706699852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3206582069396973, "target_z": 1.5703125134048492, "ce": 6.038498878479004, "genuineness": 4.193545341491699, "identity": 0.7501996159553528, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/77428683b2e8871c1210034227de5e058740e87212deb1def55114831949934e.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": "774a40a75fa4637a280e8f908935c10bf2981436ef405866661dd80391faea7b", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63fdac7282dd6d7932f22d5c04aa860b2cf3fa2a3ef961682493e55362c7790d", "generation_seconds_batch": 22.71756419702433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.8131817579269409, "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_VOLT_reg_low/774a40a75fa4637a280e8f908935c10bf2981436ef405866661dd80391faea7b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "77869a57a9607f958bc776d9748882c9ec16366d9dbc129df61eb7b51261b367", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85b2af8395bf7c7430177dda419e7dd0a0f48f76f0190526cb0e7dfe1d528fa4", "generation_seconds_batch": 18.04255881602876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.740500807762146, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/77869a57a9607f958bc776d9748882c9ec16366d9dbc129df61eb7b51261b367.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "8f3fc85732455800e536ad57a4f1df3f673f2be13828f4b0514bc553f1fc7405", "generation_seconds_batch": 15.187313472968526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.1894097328186035, "genuineness": 4.6705241203308105, "identity": 0.7487938404083252, "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_VOLT_reg_low/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "77c1e345d44e79df789728503cab7296756c22130dbc70cfd4726ddadaa3c52e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "23687c92f8fb3538be0f2ea8e89ae9a675b7be21424a3794cbbac2f02f1abe21", "generation_seconds_batch": 21.659675510018133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.046441555023193, "genuineness": 4.168440818786621, "identity": 0.7414876818656921, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hace.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/77c1e345d44e79df789728503cab7296756c22130dbc70cfd4726ddadaa3c52e.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "77e6cec9c455525152f6cdd9a5b3b7b60dd9c4d2f511ccc0488ac0d5dac9da22", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ed5fababc1f3e0f5e2000f2f146442be9478e18642161310ebc02096a579849e", "generation_seconds_batch": 21.16075615095906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1713922023773193, "target_z": 1.7421875080429094, "ce": 6.197352409362793, "genuineness": 4.519899368286133, "identity": 0.8116838335990906, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/77e6cec9c455525152f6cdd9a5b3b7b60dd9c4d2f511ccc0488ac0d5dac9da22.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "78295dc7791f4eb271f8ad4f4de70a9742ce8b2197880d30076cd133ca566820", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aaf8a6a1e6e56fcac400ca1601b60e692351b0568721321f8a2ee755871afae1", "generation_seconds_batch": 25.937193818041123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 6.149696350097656, "genuineness": 4.218649387359619, "identity": 0.7567098140716553, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/78295dc7791f4eb271f8ad4f4de70a9742ce8b2197880d30076cd133ca566820.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "784ab972c1281a10ec599064e04f4e2119069a9fb20b5b3ce272b5c6df544161", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c6422f0076babe4547d840f6dd9178d5ee37101ce5cff45f1023e438c26de6e", "generation_seconds_batch": 15.555638399906456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.632759928703308, "target_z": 1.2109374372653057, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.7316559553146362, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/784ab972c1281a10ec599064e04f4e2119069a9fb20b5b3ce272b5c6df544161.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "78908cf624a650fd042484c1f2465e6c7c7d3c781e433acb662e1ba94845659e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c18e592e0c095dfcbd17bddd0bf1eb0354e57d7c4657d65819b747210615694", "generation_seconds_batch": 21.51033430895768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 5.998785495758057, "genuineness": 4.319066047668457, "identity": 0.7135355472564697, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/78908cf624a650fd042484c1f2465e6c7c7d3c781e433acb662e1ba94845659e.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": "789926d4a04c82cdaf5e682f7270db538413f91693b7f24a2bac1a804e6d4c7e", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fffe5c277c9b2e324ec0541597822a7609ac2343131c63791064442c9ca09179", "generation_seconds_batch": 15.835111622000113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.1894097328186035, "genuineness": 4.400654315948486, "identity": 0.7081040143966675, "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_VOLT_reg_low/789926d4a04c82cdaf5e682f7270db538413f91693b7f24a2bac1a804e6d4c7e.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "78d492d43d6522737df18775d443884a079c8058e06ab4d0c0329cbeb3c1589a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76d18a6b68b07c13078f7684ce725c964152714ebb7de0e053cdfc28539b8f9e", "generation_seconds_batch": 16.79067621799186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.268836498260498, "genuineness": 4.507347106933594, "identity": 0.6480129957199097, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/78d492d43d6522737df18775d443884a079c8058e06ab4d0c0329cbeb3c1589a.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "790609c5661058d92a1d2d4451fe95e4864d770e56e4d0c83eb621f9fbb8f0e9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce15cd1b87c45e655691fc5104723bf6de429342587503afd37dc08fa79cd254", "generation_seconds_batch": 22.470468753017485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6802536249160767, "target_z": 1.1562499764074654, "ce": 5.982900142669678, "genuineness": 4.149612903594971, "identity": 0.6177196502685547, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/790609c5661058d92a1d2d4451fe95e4864d770e56e4d0c83eb621f9fbb8f0e9.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": "7920b7769c59ea0733607fe890db951a548946751217941a4d289d8050e5c2d9", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4ad642b2195907fb7d63972110f09ee89edcadf47a6eca7a7611e9c411434ac", "generation_seconds_batch": 21.180753358989023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.743329644203186, "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_VOLT_reg_low/7920b7769c59ea0733607fe890db951a548946751217941a4d289d8050e5c2d9.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "7920c81068d307d8fff524ecae4890d81863ca8470c4a92ff53695fee1193a71", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf8af15e442b18548922216704c21b1a4e9309e3179a5d716f02f32d2fc40647", "generation_seconds_batch": 15.8695157780312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088608026504517, "target_z": 1.468750066488052, "ce": 6.205295085906982, "genuineness": 4.362998008728027, "identity": 0.5093938112258911, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7920c81068d307d8fff524ecae4890d81863ca8470c4a92ff53695fee1193a71.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "79254a815be5f368de20576ea1cc26873404daaf2c1649d7c26609842737ce1f", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "515652aaa778a0bf572e64db714b63a35a2c370d1f6d58c9375469121c528d49", "generation_seconds_batch": 15.61982711602468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5445572137832642, "target_z": 1.3125000214477587, "ce": 6.348263263702393, "genuineness": 4.444586753845215, "identity": 0.3585112690925598, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/79254a815be5f368de20576ea1cc26873404daaf2c1649d7c26609842737ce1f.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "cd6ed2d9d3cd430bd78100eff1ab2929db4190cdc868f928b03e1c29338b5329", "generation_seconds_batch": 15.95900203997735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.165581703186035, "genuineness": 4.469690799713135, "identity": 0.6969676613807678, "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_VOLT_reg_low/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.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": "7961b7b871db3fd7da0d44df26239925e852e22d878cd142eaee4cf8e6a216a9", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bb23670a40d3052baddc0d3a9ae2a4b000c11161df6dfdc9eeb02528a2f4b8f", "generation_seconds_batch": 22.346272171009332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.022613525390625, "genuineness": 4.306513786315918, "identity": 0.8185797929763794, "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_VOLT_reg_low/7961b7b871db3fd7da0d44df26239925e852e22d878cd142eaee4cf8e6a216a9.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "796d765e655e584e9a8ae5be3bf660fef29828c6bb6e79365703b09168e620e5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59a4d608da8024d0a10747ea2dd966aed913a2ef21f840ddd9fe6e34dcdebc0f", "generation_seconds_batch": 21.42492289992515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.252951145172119, "genuineness": 3.9173994064331055, "identity": 0.3466288447380066, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/796d765e655e584e9a8ae5be3bf660fef29828c6bb6e79365703b09168e620e5.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "797fd52c5b2f2c1b795dca605f20e0fbe5446ee5d08fa3f218de224f8d39dca1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "52f0577e25b51347c5a5605e5bf1edce5edc1ac904224b47cd4bba85348bc057", "generation_seconds_batch": 16.583072017063387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.2767791748046875, "genuineness": 4.381826400756836, "identity": 0.5440682768821716, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in keihard.", "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_VOLT_reg_low/797fd52c5b2f2c1b795dca605f20e0fbe5446ee5d08fa3f218de224f8d39dca1.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "7a61800c3b14b17627ca58261fe133f40e2455b1aca299ada98fbc1f04ded0df", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b20c333c95f00ffc0e32b1e099d4f658cda1cf7ffc39f6ad2e49b72f5647e40b", "generation_seconds_batch": 22.565550635918044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7032396793365479, "target_z": 2.281250053619397, "ce": 5.903473377227783, "genuineness": 4.507347106933594, "identity": 0.7883472442626953, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7a61800c3b14b17627ca58261fe133f40e2455b1aca299ada98fbc1f04ded0df.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "7b01297a737c3bcc36c604da345d86486a9432039966b058c58a4f1de9a30d7f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f21ff868f3738054173452fa2409965cc81dfd3ff7092060c31de2cbd923c09a", "generation_seconds_batch": 16.375400315970182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.0146708488464355, "genuineness": 4.6203155517578125, "identity": 0.7013880014419556, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7b01297a737c3bcc36c604da345d86486a9432039966b058c58a4f1de9a30d7f.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": "7b29a2517ec2d5fc796f11712be3c2bb36075ccd1ea94655f39871d2c9463cf9", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "932fe13043cf1ba08588b963bbb2a4e32cce7b75467f74e05c906aae0581f8bb", "generation_seconds_batch": 19.92068823997397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.3641486167907715, "genuineness": 4.356721878051758, "identity": 0.19548514485359192, "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_VOLT_reg_low/7b29a2517ec2d5fc796f11712be3c2bb36075ccd1ea94655f39871d2c9463cf9.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "7bae54997975ff78b841021c2e1a3b71753b9a2cab0f954ef852994e9eae528c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "4f5e39d17401f48bf17f292260b72be3697602ebbc3f91cb44c5fcd386b60254", "generation_seconds_batch": 22.562952280044556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 5.927301406860352, "genuineness": 4.2751336097717285, "identity": 0.6445868015289307, "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_VOLT_reg_low/7bae54997975ff78b841021c2e1a3b71753b9a2cab0f954ef852994e9eae528c.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "7c4e023c37a327bcd62771421a8efd7d8e1e1857ab07dbfd12acce2ecf83a591", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "985c1fede33afb35ca4fd9c35fcf510f7099314cbfeef489e8a9c71c2910178b", "generation_seconds_batch": 21.223842097097076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7303789854049683, "target_z": 2.250000017158207, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.7830122113227844, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7c4e023c37a327bcd62771421a8efd7d8e1e1857ab07dbfd12acce2ecf83a591.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "7c8a12fbd1557396895d1efb14b4de3aa7677fcbf29cc9b10992ccdc967dd4e6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b65a7dab072f6b30704ecc9122efe7909240af0e13c6ae8b0101a08ff17e7df0", "generation_seconds_batch": 21.92227791901678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.8038758635520935, "hypothesis": "La porte 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_VOLT_reg_low/7c8a12fbd1557396895d1efb14b4de3aa7677fcbf29cc9b10992ccdc967dd4e6.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "7d7509e7a017ecad7151fe26e97b294fbc32c6fa24bcd1e19e29e6a2986801c8", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdea90fd652c5bef5ef5390935190a39a84d9f9905c8b9481f591463f1a998d9", "generation_seconds_batch": 15.188094401033595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.252810001373291, "target_z": 1.6484375359249959, "ce": 6.165581703186035, "genuineness": 4.444586753845215, "identity": 0.6558972001075745, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7d7509e7a017ecad7151fe26e97b294fbc32c6fa24bcd1e19e29e6a2986801c8.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "7d7c9d1105c088fcf9fea23b124b6a65b4e27fa49d9a780e0503a9c07d518ee1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db0594495a24f5ab38f0c89d9e63befb192cc672d1d0b8a969eb58d3b35adab9", "generation_seconds_batch": 16.544953971984796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.2767791748046875, "genuineness": 4.570107460021973, "identity": 0.6782073974609375, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7d7c9d1105c088fcf9fea23b124b6a65b4e27fa49d9a780e0503a9c07d518ee1.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "7d9081ac1dbd22eab96012a54ea7a35f6e36e65f8c1d79eeb22424229fb44570", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92dddddec278d8589f22a373e7464d7a02083ab09a63d442c66c96e0c8b64b0e", "generation_seconds_batch": 16.02732338407077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 5.919358730316162, "genuineness": 4.457139015197754, "identity": 0.7295474410057068, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7d9081ac1dbd22eab96012a54ea7a35f6e36e65f8c1d79eeb22424229fb44570.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "7dbd4942b0263766210277815b6d951e9ba81485c235c0c1665532889dad3a43", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40aa56896e8b9b66e5de27d281db7d0ffa16be08ad4cb55c1263d96118c59ddd", "generation_seconds_batch": 20.276484642061405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.340320587158203, "genuineness": 4.256305694580078, "identity": 0.4370972514152527, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7dbd4942b0263766210277815b6d951e9ba81485c235c0c1665532889dad3a43.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": "7dbf8ddd00a7d1080fd6c5f63c726e476f85afb120fb2b37a4274d78ea50e162", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d86d083c97e312259608ad6aab24a7fd6b06069e19eaa629a4faa857b643445", "generation_seconds_batch": 14.814975772984326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.157639026641846, "genuineness": 4.231201171875, "identity": 0.6933643817901611, "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_VOLT_reg_low/7dbf8ddd00a7d1080fd6c5f63c726e476f85afb120fb2b37a4274d78ea50e162.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "7dfbbbe26cbf1d221ba74de56ed2a2b07025933e4ec0eb09fd28d1b1f4845d46", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a828ba4bfec3d1645b4e785a85db653c0bee8dd45d53a1b265c3ee6be667f329", "generation_seconds_batch": 21.220728573040105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.372091293334961, "genuineness": 4.287685871124268, "identity": 0.2740134298801422, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7dfbbbe26cbf1d221ba74de56ed2a2b07025933e4ec0eb09fd28d1b1f4845d46.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "7e2702d7b0c408465777d3f19b76dc783dfd8cfa82e949d6e16fbbef7e6dcb5a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f00ea3dbafd58ca2627db230c7cd3df99c58b29bc96d11bff4102e4db780cef", "generation_seconds_batch": 21.437006674939767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.23706579208374, "genuineness": 4.218649387359619, "identity": 0.37075725197792053, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7e2702d7b0c408465777d3f19b76dc783dfd8cfa82e949d6e16fbbef7e6dcb5a.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": "7e3dfcb8b511aefa26ed96b60b7f8a84e44c84eac24e6bb8a4b4a370d8af48e4", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c195ba26442f96a767ded5db23750c9f2374e69592aff41f00b02d77dd7240c", "generation_seconds_batch": 22.390626218984835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.0146708488464355, "genuineness": 3.9927117824554443, "identity": 0.839964747428894, "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_VOLT_reg_low/7e3dfcb8b511aefa26ed96b60b7f8a84e44c84eac24e6bb8a4b4a370d8af48e4.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "7e646ca052b729ec973ca61cda0667c882612105f488bef800ea0ff8bf1f843c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2b3195ef081158f64fc4b0606650009f23c08200ac6bc2acd05f70c96ba587c", "generation_seconds_batch": 20.821451934985816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.402076005935669, "target_z": 1.4765625412869354, "ce": 6.117925643920898, "genuineness": 4.532451152801514, "identity": 0.7084357738494873, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7e646ca052b729ec973ca61cda0667c882612105f488bef800ea0ff8bf1f843c.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "7e99dc73931780be288af551b10d2163d7645302f572e44aedb67111926968aa", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58f6292f6ead6308ee7009334cf2f013a4b42c5a2a66e47aefd6db84eae70ffb", "generation_seconds_batch": 21.92227791901678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.8006315231323242, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7e99dc73931780be288af551b10d2163d7645302f572e44aedb67111926968aa.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "7eb1479ea269f5c46b56ef71ee681d5f93dab0cd4c4d68ee98b64b78a7685717", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "580f0d622a297f6c805b54d388f2a3d44c3e321ac55b27e8cf62fd05296c752c", "generation_seconds_batch": 21.434035498998128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0628350973129272, "target_z": 1.867187516622013, "ce": 6.173524379730225, "genuineness": 4.281409740447998, "identity": 0.6613067984580994, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7eb1479ea269f5c46b56ef71ee681d5f93dab0cd4c4d68ee98b64b78a7685717.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "7f13e275f3e2941339e687f300c57aa6779be9f936cf89a00b4da773aafb8c95", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41835f51f8e13ae4a729d29d59103ff3f25b96c56e97710dd5ff28f996e70d4c", "generation_seconds_batch": 20.505616639042273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.086154937744141, "genuineness": 4.042920112609863, "identity": 0.6888003349304199, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7f13e275f3e2941339e687f300c57aa6779be9f936cf89a00b4da773aafb8c95.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "7f2eb10d6474cdfbf192ab651f050de5aadc5b3be6c50632ed65351bdd1ec532", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a31870abbc986d8b6895e66cfe60509af762ae8f3442bced8e40177a63ad4246", "generation_seconds_batch": 22.75454349303618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3817216157913208, "target_z": 1.499999965683586, "ce": 6.109982967376709, "genuineness": 4.017816066741943, "identity": 0.6592015624046326, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7f2eb10d6474cdfbf192ab651f050de5aadc5b3be6c50632ed65351bdd1ec532.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "0764a0d3946bfd752c9774daa8c36eb25f9494e757f470d159a93efed30dca22", "generation_seconds_batch": 17.113209042930976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.348263263702393, "genuineness": 4.457139015197754, "identity": 0.2250417023897171, "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_VOLT_reg_low/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "7fb8dd43860f3f10d049a321eaac33b93479988d1c97858e0044ddf2cb4f1fa0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdde121e3d1903673dca361a65d8401885890d0b4764492729962054710bcedf", "generation_seconds_batch": 23.154373478959315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.6101636290550232, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7fb8dd43860f3f10d049a321eaac33b93479988d1c97858e0044ddf2cb4f1fa0.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "7fe18e02cc05c8a84f70b03cf62db9eccdaf15b79cfb2f8ac608a4ec254f9d3c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c137bdece5185d12e787cbd15814970c72a5c5502d65e853dd3d2814f0002a8", "generation_seconds_batch": 16.77278904907871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.6970378160476685, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/7fe18e02cc05c8a84f70b03cf62db9eccdaf15b79cfb2f8ac608a4ec254f9d3c.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "8011e212b806b91a89ca6bd3aafb28fdd20d5bd1abe98f9e5c8480692ee4c062", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6fa6c9c571d2f4705f226307bfc89304e2aeef4ff2cfa3bf68384ed584613c4", "generation_seconds_batch": 21.071168583934195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6625308990478516, "target_z": 2.3281249024126978, "ce": 6.038498878479004, "genuineness": 4.419482707977295, "identity": 0.8904992938041687, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8011e212b806b91a89ca6bd3aafb28fdd20d5bd1abe98f9e5c8480692ee4c062.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "801946f6da46afd03dfa85a807c1986abfcc750a774362457cdf37acaac1424e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "862d638dfcdd2809358624303def32b3d41003a602b890c6b99d9c3f97facd4d", "generation_seconds_batch": 20.84800118894782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.284721851348877, "genuineness": 4.400654315948486, "identity": 0.571571946144104, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/801946f6da46afd03dfa85a807c1986abfcc750a774362457cdf37acaac1424e.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "80227eb9e4998368d7763d4952aaee2abd7e6b1066f9bfe591c60129f245bec0", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04307989680f9f7559980ebd8536b2986fd0db74a80ce5a13da80894d9fa27b2", "generation_seconds_batch": 15.055178398964927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.229123115539551, "genuineness": 3.8295347690582275, "identity": 0.38321948051452637, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/80227eb9e4998368d7763d4952aaee2abd7e6b1066f9bfe591c60129f245bec0.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": "80761b70e06b85a45b410aa8cde5b04b9cdadc7f41fc612fc05a47b4c4360f52", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ee3b3a6651b5ebb1eb16a6412a35205c0158934043799d37be3dec26b3024ef", "generation_seconds_batch": 15.487331610987894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.403861999511719, "genuineness": 4.325342178344727, "identity": 0.4207659065723419, "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_VOLT_reg_low/80761b70e06b85a45b410aa8cde5b04b9cdadc7f41fc612fc05a47b4c4360f52.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "8085c9eeb5256034fa2e8450a562a4e5407eacb673dd0aa9976c3da38e84e06d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "496f82e69719c728aa814f9c92aa645ec6f558033caeaa342537792753eb9680", "generation_seconds_batch": 19.56816331099253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.24500846862793, "genuineness": 4.18726921081543, "identity": 0.7058579325675964, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8085c9eeb5256034fa2e8450a562a4e5407eacb673dd0aa9976c3da38e84e06d.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "808efa76d94a7d55dda40dc321fd60d054ed777b033263cb273ff3628c1a5a62", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7f80ffc81dabba16b70672aac1e0083090f46278806961208d669a53036d6760", "generation_seconds_batch": 21.76626543502789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.648961067199707, "target_z": 2.3437501265417766, "ce": 6.229123115539551, "genuineness": 4.419482707977295, "identity": 0.8195439577102661, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in keiten.", "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_VOLT_reg_low/808efa76d94a7d55dda40dc321fd60d054ed777b033263cb273ff3628c1a5a62.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "546f2513a2515168e593e7eb02370a953f4ab12c7d8ab0c8746079c0c974c3c5", "generation_seconds_batch": 15.186123343999498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.0146708488464355, "genuineness": 4.482243061065674, "identity": 0.6811721324920654, "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_VOLT_reg_low/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "80d082335cbe4c5650146074307b7f80ed961009d4691df9cbe8b9dc585e64cb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2245bd80a044df747245a4a57be8e20187d4a984c408856451ee3f7e96c1eac3", "generation_seconds_batch": 22.562952280044556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.09409761428833, "genuineness": 4.331617832183838, "identity": 0.6622005105018616, "hypothesis": "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_VOLT_reg_low/80d082335cbe4c5650146074307b7f80ed961009d4691df9cbe8b9dc585e64cb.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "80ecea85bc68c799af5acddc1c22f1c25e43455ec0eff954dccbb13322c27ee0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "475ee47b3fea37557f1e89cccce0dffcbf3a5ac48140c11effd50859c5e763ec", "generation_seconds_batch": 22.79530879703816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.994986891746521, "target_z": 1.9453125391421597, "ce": 4.239483833312988, "genuineness": 4.6705241203308105, "identity": 0.7238247394561768, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/80ecea85bc68c799af5acddc1c22f1c25e43455ec0eff954dccbb13322c27ee0.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": "81381944ee0c81732e50b513fbaa2087d0491bb304e65dd2aa7065e56f163314", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66ad1f0b3c94638de992fea90b6e0a3e4078d7025e7dd8f6f7c0afd04b957e85", "generation_seconds_batch": 19.92068823997397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.435632705688477, "genuineness": 4.137060642242432, "identity": 0.30826714634895325, "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_VOLT_reg_low/81381944ee0c81732e50b513fbaa2087d0491bb304e65dd2aa7065e56f163314.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "81729a96112f8add35c7f4f973df7faca3fa69444f4caa7b85623fe80e3183b8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7757b377d8a08258cf12fff91e00f0e657508c6065a0e096a15deedc2c4396a5", "generation_seconds_batch": 22.77604750590399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.022613525390625, "genuineness": 4.494794845581055, "identity": 0.5199769735336304, "hypothesis": "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_VOLT_reg_low/81729a96112f8add35c7f4f973df7faca3fa69444f4caa7b85623fe80e3183b8.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "81bf0cef1bc212bef2323b596814127aeaa9dd6d223591d2e053ba7171e97f26", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7747065c586b03fe3510aa50ff493cd1d0f775f8e6394ca53177c842e056706c", "generation_seconds_batch": 15.90894788398873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5038483142852783, "target_z": 1.3593750075067155, "ce": 6.340320587158203, "genuineness": 4.532451152801514, "identity": 0.7165800333023071, "hypothesis": "Ik 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_VOLT_reg_low/81bf0cef1bc212bef2323b596814127aeaa9dd6d223591d2e053ba7171e97f26.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "81c2ff144098a3a8d56d0d453492821b4bc1c447b922dd9c4a0696e5157ba558", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce01f2523a5c392a80b027ee54a061e03ee8322209ed8f786d1cc9c038f3e9f4", "generation_seconds_batch": 14.62107526208274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.252951145172119, "genuineness": 4.494794845581055, "identity": 0.7193655967712402, "hypothesis": "Que he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/81c2ff144098a3a8d56d0d453492821b4bc1c447b922dd9c4a0696e5157ba558.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "82271d42eeebcf534ee3bee34ea5b62d01ed5dea85e160f1e01f2bbfe710007e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6ef719bbc920ca513cf23078767868a8fdc510dfa7d264e8bc676500943c178f", "generation_seconds_batch": 22.148645624984056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7982271909713745, "target_z": 2.17187499463806, "ce": 5.871703147888184, "genuineness": 4.337893962860107, "identity": 0.7870640158653259, "hypothesis": "He que he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.35294117647058826, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/82271d42eeebcf534ee3bee34ea5b62d01ed5dea85e160f1e01f2bbfe710007e.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "823e52700b66e565cf039f234888b1b7f8c4f8f2e1712abf0372e8a47b1a5746", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ef3037cd92319043bc576bdb36ab7f344da4b54dd4e8c97b36ecc8f56981138", "generation_seconds_batch": 16.147964627016336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.062326908111572, "genuineness": 3.8546390533447266, "identity": 0.628600537776947, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/823e52700b66e565cf039f234888b1b7f8c4f8f2e1712abf0372e8a47b1a5746.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "824f1083465491126c3f037de5954f31854599df83de2c50ed48d6b56dd835e5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75520e926ca046d2722955aee7289d448d187a1478e51c6f9976df6e90a93fdd", "generation_seconds_batch": 22.438938518054783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.459460258483887, "genuineness": 4.174716949462891, "identity": 0.5966895818710327, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/824f1083465491126c3f037de5954f31854599df83de2c50ed48d6b56dd835e5.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "829353d16eccd73bb32349a4f4f715855e98ee60c341a29c5c71e39a183267f8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72b8f9fb3104516ecc44ac4a708e5bea4d5c823d7e011c705fa1f0387e01dfb8", "generation_seconds_batch": 21.329932671040297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.427689552307129, "genuineness": 4.381826400756836, "identity": 0.3485161364078522, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/829353d16eccd73bb32349a4f4f715855e98ee60c341a29c5c71e39a183267f8.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "82ba1ee787cf0d2c99f2339299beb2396935bfc319f3ce35f9d49458613d2b0c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3627894e6575920b1839865bf3adf2c036a64fd26d31a2bb935e58bd1c8c4008", "generation_seconds_batch": 22.624964650021866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7710878849029541, "target_z": 2.2031250310992503, "ce": 6.165581703186035, "genuineness": 3.6789097785949707, "identity": 0.4091421365737915, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/82ba1ee787cf0d2c99f2339299beb2396935bfc319f3ce35f9d49458613d2b0c.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "82f2449743d3e7a52da7827e011d476ed64872cc302903b7131ad67200473633", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb08656d57179ffd5ed553d573c07ce8d391453719513098e45230bebb1649d4", "generation_seconds_batch": 18.894496402004734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 6.09409761428833, "genuineness": 4.2249250411987305, "identity": 0.4323868155479431, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/82f2449743d3e7a52da7827e011d476ed64872cc302903b7131ad67200473633.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "82fb34d4de4bcc4fd84225cb951f895b0d33e8fe904ee8a270562ce1576d02d0", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3aaecf33161758f9069d59ea67f4609a1d98d407ffb526891e1124f2aa9786c6", "generation_seconds_batch": 15.741016366053373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.2767791748046875, "genuineness": 4.017816066741943, "identity": 0.1904647946357727, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/82fb34d4de4bcc4fd84225cb951f895b0d33e8fe904ee8a270562ce1576d02d0.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "8363dd229e728b2c4e8c5ee42d8a129f81efc9cbd35fea87f350f9673c8f3ad7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd39e1db64a818d81b94a8c5992fe15a995afeb65d79ec35805ae87779e04408", "generation_seconds_batch": 19.696258637937717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.268836498260498, "genuineness": 4.268857479095459, "identity": 0.6909947395324707, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8363dd229e728b2c4e8c5ee42d8a129f81efc9cbd35fea87f350f9673c8f3ad7.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "839ad4c3f712ae6063a062e2feac9f7a8a1bceac3f343be611801dad24709607", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6273325a1a9332ec0040dd81ecc33d2dfc9bdb69dc92d0ea34f86894447bf29", "generation_seconds_batch": 21.61740483203903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.308549880981445, "genuineness": 3.911123275756836, "identity": 0.6260204315185547, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/839ad4c3f712ae6063a062e2feac9f7a8a1bceac3f343be611801dad24709607.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "8412b09bc9c4f6456642b505bfabdf223f2145789f44bd44823f5e0067088f14", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "99b0e9412ed1c59d827db472d4717687a5b6fee8703a07125143ef18f50209b4", "generation_seconds_batch": 22.29804027301725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.284721851348877, "genuineness": 4.143336772918701, "identity": 0.6178808212280273, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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_VOLT_reg_low/8412b09bc9c4f6456642b505bfabdf223f2145789f44bd44823f5e0067088f14.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "84bf971c9a675b3e0ac45aa5239cd227581622041b47a27c6fbf004c0a4c0824", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0677bdc2cfa737aa7c718d2d4310cabaf15c4153075515b3431092cd1d14d3c6", "generation_seconds_batch": 26.704046608065255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.103543996810913, "target_z": 1.820312530563056, "ce": 6.022613525390625, "genuineness": 3.8420867919921875, "identity": 0.5133287906646729, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/84bf971c9a675b3e0ac45aa5239cd227581622041b47a27c6fbf004c0a4c0824.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "85054e71f1db980e0f44908aabc257ac303834057dde7fdcc4abbc50c5d0097d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f833d275c0e8136c0916bdc53bc0f731e31a53d9b1617c1961e53a45bf1526b", "generation_seconds_batch": 21.46703628601972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.054384231567383, "genuineness": 4.061748504638672, "identity": 0.6299906969070435, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/85054e71f1db980e0f44908aabc257ac303834057dde7fdcc4abbc50c5d0097d.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "855d5ec8e83aa4d7016c1a76d989f5974aa94e104856bafe17e1a636cf5b4012", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a38b3a1cc82cfa1660678f283ed32c4aa0208c5eaf62f72ea06016b89ed46ae", "generation_seconds_batch": 22.038292304961942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.510633111000061, "target_z": 1.351562532707832, "ce": 6.070269584655762, "genuineness": 4.193545341491699, "identity": 0.23691189289093018, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/855d5ec8e83aa4d7016c1a76d989f5974aa94e104856bafe17e1a636cf5b4012.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "85cea810e4950ca32810521dde4034a17352d59738d8bc19f72a0db420b5fd8f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2014f63bdaec01036c9ea91629de0f48b55e9dd9e458f2ac6f8457a2b727a9c0", "generation_seconds_batch": 23.257234496995807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 6.1894097328186035, "genuineness": 4.074300289154053, "identity": 0.6266151666641235, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/85cea810e4950ca32810521dde4034a17352d59738d8bc19f72a0db420b5fd8f.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "85ec544f4f27dfc751ee8bb0ab17d081efdb81b3c251c6f45b622b9074247747", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4926ce3c556a7b5ebab0e96f2c000a00a5bac9ad07d04e85e784d530bd67d0b", "generation_seconds_batch": 22.55817770701833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.006728172302246, "genuineness": 3.773050308227539, "identity": 0.5647006034851074, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/85ec544f4f27dfc751ee8bb0ab17d081efdb81b3c251c6f45b622b9074247747.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "868c52f44c324636058dd7918fbf2704b93c6964bbc1c1c7300e8211d2c62390", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7d0bd44db5a62d0c6a479740f9884f789bb854cb89ed30916f5164ae1b601eb", "generation_seconds_batch": 22.176205517025664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.300607204437256, "genuineness": 3.8420867919921875, "identity": 0.29641789197921753, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/868c52f44c324636058dd7918fbf2704b93c6964bbc1c1c7300e8211d2c62390.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": "86f33165c5348ab0d60f28f425cc78e067d1a82beaed7f5f0c2b9e20a8fe6121", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Someone moved the little blue chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c38fb059aa1ad7e05c95bc478f901efd472069b72dd5eb14388e00880da1280", "generation_seconds_batch": 15.254258602042682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.419747352600098, "genuineness": 4.419482707977295, "identity": 0.276176393032074, "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_VOLT_reg_low/86f33165c5348ab0d60f28f425cc78e067d1a82beaed7f5f0c2b9e20a8fe6121.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "870596ae48d65366b95dffa80776ede168f77be82be460425a4b0f20931e58cd", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f64f116eb6e1a2efcbdfd07fd0869c4256a37a55ab196371966c47c090240bd7", "generation_seconds_batch": 15.34919295005966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.449569821357727, "target_z": 1.4218749431634394, "ce": 6.459460258483887, "genuineness": 4.344170093536377, "identity": 0.5467134714126587, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/870596ae48d65366b95dffa80776ede168f77be82be460425a4b0f20931e58cd.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "87128578fb894393e749120f95d2f33a61dc0ea9a416da3fa05672aff2273569", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c3b9f709d90492d27c20ebdaaa145112209519ef56a25060cd8e1efa2071743", "generation_seconds_batch": 22.306950291036628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.221180438995361, "genuineness": 4.218649387359619, "identity": 0.3151186406612396, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/87128578fb894393e749120f95d2f33a61dc0ea9a416da3fa05672aff2273569.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "87764857e0f01652aee75b879dbc34750ce59cbe0d332651c4d2d58101ba7397", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27912681776516a00ff2889358455baf863693cc103011079711b3933e8d5da1", "generation_seconds_batch": 22.294962006038986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.427689552307129, "genuineness": 4.375550270080566, "identity": 0.24668067693710327, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/87764857e0f01652aee75b879dbc34750ce59cbe0d332651c4d2d58101ba7397.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "87a41b50eccf17d0124188f941ad0584ea54457d6eb61eccf02ada1330fb032d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a44e3ef2389e98779d0bbf76eaf93cde343cfe4a28f4462086e61b68124be0d", "generation_seconds_batch": 22.294962006038986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.38797664642334, "genuineness": 4.532451152801514, "identity": 0.40341806411743164, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/87a41b50eccf17d0124188f941ad0584ea54457d6eb61eccf02ada1330fb032d.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "881509cccb4d19864aa809ab21194776eba6c2ea4cee915a317acc8a2dab851e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "374aad2d93fc8319ed0c4ce56d30458b84b1f408261e23e96318c7d33be50a2e", "generation_seconds_batch": 18.05265281291213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 5.772419452667236, "genuineness": 4.18099308013916, "identity": 0.7003694772720337, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/881509cccb4d19864aa809ab21194776eba6c2ea4cee915a317acc8a2dab851e.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "88e64a6fdfb9a2021c047db14e83a428bd68424d420ed3f15b77c1ab2964f264", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30f7a179d2eac6b8de1d83b04afa54466d32a04be917115ca59af4fb59a87e50", "generation_seconds_batch": 21.62253045500256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.008556604385376, "target_z": 1.9296874522787368, "ce": 6.205295085906982, "genuineness": 4.425758361816406, "identity": 0.7107870578765869, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/88e64a6fdfb9a2021c047db14e83a428bd68424d420ed3f15b77c1ab2964f264.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "898a03d567c22877995f6b8d12eaca56577f9d5b58f0c8f9954de47ac1363af2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51c00af2b5a4c6e288d298b4b666a442aa40d26f38bd280dea96b697f1fe9195", "generation_seconds_batch": 23.02227972599212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.648961067199707, "target_z": 2.3437501265417766, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.5916976928710938, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in keer.", "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_VOLT_reg_low/898a03d567c22877995f6b8d12eaca56577f9d5b58f0c8f9954de47ac1363af2.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": "89e0024d19b0aba55264e5761ece9d60a183dd13825a017339f901aacf2cd363", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27a9ccca40ebffbf651e064782213f42114ec34a562bb8e5e41252216e3121f5", "generation_seconds_batch": 16.085609073983505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.109982967376709, "genuineness": 3.9425034523010254, "identity": 0.7417842149734497, "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_VOLT_reg_low/89e0024d19b0aba55264e5761ece9d60a183dd13825a017339f901aacf2cd363.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": "8b597a76b05dbc4709eaacf724e876207d67a466eac6fdd1f8c08cfeaab41841", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2db1cb0c2a7b478ecfe3058e3b49a0e3c16ff94906269ba7d9ff09e117a3c06", "generation_seconds_batch": 15.875433427048847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.419747352600098, "genuineness": 4.507347106933594, "identity": 0.6429174542427063, "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_VOLT_reg_low/8b597a76b05dbc4709eaacf724e876207d67a466eac6fdd1f8c08cfeaab41841.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "8baf4c996357057ad92966a81b5dd5c7ad1bafe99fd44cd6d11815352a42ae23", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "011dba480de772b58a4d7999f457a064a81dd7cf3e3d4ef9fc8417cce1a0b8aa", "generation_seconds_batch": 16.143575005000457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 5.895530700683594, "genuineness": 3.8546390533447266, "identity": 0.6737548112869263, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8baf4c996357057ad92966a81b5dd5c7ad1bafe99fd44cd6d11815352a42ae23.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "8bdbcfa4caff3ba06526316bb4ff6a9aac62e763396e90b88322c6f2015d80bd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0ecef08cae80991a69ebe379f1b85f29838b6b5ec7b39505d271a224baa31a1", "generation_seconds_batch": 15.140744079020806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.7389000654220581, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8bdbcfa4caff3ba06526316bb4ff6a9aac62e763396e90b88322c6f2015d80bd.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "8c005e4a579614d30f489931e286304f3a549a9fcf11d66d552d3ac355bbb799", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b5a1a186e6b3add6f7507c80c6e078bb8047e9974152931eb44e6a9da7438a6", "generation_seconds_batch": 15.61982711602468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5038483142852783, "target_z": 1.3593750075067155, "ce": 6.268836498260498, "genuineness": 4.469690799713135, "identity": 0.3847460448741913, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8c005e4a579614d30f489931e286304f3a549a9fcf11d66d552d3ac355bbb799.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "8c6c68a73fb3fde3a28812cb82ea303d9fcdf4b061bbb3af937bb4ce801662f7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7621e3ceac3e1d531394a64e629315b27fc7b298b5d9974afd518fb599dd8994", "generation_seconds_batch": 24.273272639955394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.348263263702393, "genuineness": 4.394378662109375, "identity": 0.7411640286445618, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8c6c68a73fb3fde3a28812cb82ea303d9fcdf4b061bbb3af937bb4ce801662f7.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "8c83f7a1fc3e9032954b23bf8a248e33d21080c0ab39364c7efdabd40c3f88fa", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "814ab02cf8085290f2ecb101273bf55b02aefe60a3ee8dd46bb370075fc21069", "generation_seconds_batch": 17.824970457004383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 6.292664527893066, "genuineness": 3.9299514293670654, "identity": 0.4911404848098755, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8c83f7a1fc3e9032954b23bf8a248e33d21080c0ab39364c7efdabd40c3f88fa.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "8cbc70bba3a13f4818276e833079d67c598983d03c4e2d8f5fc1e931fa7365e6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5dcf050d755a07d6612b73426c5fad8c01ebc599c7be4d291d61fb3c356a5c7", "generation_seconds_batch": 15.527029410004616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.8166009187698364, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8cbc70bba3a13f4818276e833079d67c598983d03c4e2d8f5fc1e931fa7365e6.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "8cbe923b3be3ee50fd31283f5b4318c12194f0492e54d125f0a99116936ac3e5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ce6238a67643b3fcb8a208589f3d7825d04c7c0e04312da96b80309f610ec6f", "generation_seconds_batch": 23.519322487991303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.268836498260498, "genuineness": 3.973883628845215, "identity": 0.5960478782653809, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8cbe923b3be3ee50fd31283f5b4318c12194f0492e54d125f0a99116936ac3e5.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "8cef44e415ceef99a38312b484f1cb87029c8f5aab5eb363aa1d2ec74c4ee56b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f988f106d2122b3fd7c9fa0c3bcad0c46f595d017ddbc320f54aee2d8e9124c", "generation_seconds_batch": 21.863088082056493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.324435234069824, "genuineness": 4.306513786315918, "identity": 0.36838725209236145, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8cef44e415ceef99a38312b484f1cb87029c8f5aab5eb363aa1d2ec74c4ee56b.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "8d0ae000c6c7a9ce59d4d7c0c1099d9171b389b201220048f0c99573a82e870c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91b32e4e6704692c5a2113e294ad9897a59678ce743d1366f4c4920c594ad839", "generation_seconds_batch": 16.334758383920416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 6.197352409362793, "genuineness": 4.111956596374512, "identity": 0.5189661979675293, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8d0ae000c6c7a9ce59d4d7c0c1099d9171b389b201220048f0c99573a82e870c.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": "8d7e7b4d146149fd4d853dd6ee74feb094ce1866344c602c7e06aac880da05a6", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The keys are on the kitchen table.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7db7715454c9c067ef43e1ea1bdfe63e989409ab00606540ebbe4b8c41fadaf6", "generation_seconds_batch": 17.552321463939734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.157822608947754, "target_z": 1.7578124576406766, "ce": 6.324435234069824, "genuineness": 4.432034492492676, "identity": 0.17117193341255188, "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_VOLT_reg_low/8d7e7b4d146149fd4d853dd6ee74feb094ce1866344c602c7e06aac880da05a6.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "8dc2c08fcc9e0b66c072b0ee5fe586aa2068f8cfe674c406a180363bbfac878e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb6187d1113111b64b180562c5d2948a204c41d2ed3832dc9bf15331ec8c9157", "generation_seconds_batch": 15.741016366053373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 6.205295085906982, "genuineness": 4.18099308013916, "identity": 0.06383053958415985, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8dc2c08fcc9e0b66c072b0ee5fe586aa2068f8cfe674c406a180363bbfac878e.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "8dd0e66674fe4ff81ffc340ddaf1965d208891ccb138255fa13efbb3cfa9af54", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bb751681006c245841f4da8eb4fc2be44b4e85fbcc2573befe217af9e23b10a", "generation_seconds_batch": 18.65719736693427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.476709008216858, "target_z": 1.3906250439679053, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.6302343606948853, "hypothesis": "He encontrado. La cajita estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8dd0e66674fe4ff81ffc340ddaf1965d208891ccb138255fa13efbb3cfa9af54.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "8e6e845fda457976cb489363a94037987598778ffd9d63851dc94c12b28dd8bf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36b08d7e4228d6fed8605d429103d63a35be68e83c559ea2da699e48a9717c29", "generation_seconds_batch": 17.66756415704731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.0305562019348145, "genuineness": 4.306513786315918, "identity": 0.7832109928131104, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8e6e845fda457976cb489363a94037987598778ffd9d63851dc94c12b28dd8bf.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "8e78316aa0cef2874581168569cbbfd677fe98915d2b892b4b452ed2f13f5710", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd663d5b7129fa6ad2743cc08a3ef79bac46f8ed732c275271c5220cac61e163", "generation_seconds_batch": 21.18937423499301, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.476709008216858, "target_z": 1.3906250439679053, "ce": 6.23706579208374, "genuineness": 4.256305694580078, "identity": 0.7291456460952759, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8e78316aa0cef2874581168569cbbfd677fe98915d2b892b4b452ed2f13f5710.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "8e8850c5535412a2ffb75eb7c770ae53fec6ba8c17274df6514f57ec21d1e1d4", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a1b4163546b853491fb5bad039959e889f21c37353233ae5901b9701a6a2abf", "generation_seconds_batch": 15.258731039939448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4834939241409302, "target_z": 1.3828124319033661, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.494344025850296, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8e8850c5535412a2ffb75eb7c770ae53fec6ba8c17274df6514f57ec21d1e1d4.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "8ef91141f2b00ccd97c6c42a6ea0d0cc7e2225b1a9b18240d45a931850f7d6e2", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97805bc2c1c29e57bfb6c35adfbcef1248f6748ddfb7d64477bb0dff68ec35b6", "generation_seconds_batch": 15.856437179958448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0628350973129272, "target_z": 1.867187516622013, "ce": 6.340320587158203, "genuineness": 4.18726921081543, "identity": 0.6806070804595947, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8ef91141f2b00ccd97c6c42a6ea0d0cc7e2225b1a9b18240d45a931850f7d6e2.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "8f08fdf6477fa0364a01234d30b4a94f7c62658fa5ce1df87a9483040ba340f8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b5a3a13fc3e059d54fdd47357dd678006faa42edbc45bd37a63535e7e01d589", "generation_seconds_batch": 25.976079149986617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1713922023773193, "target_z": 1.7421875080429094, "ce": 6.125868320465088, "genuineness": 4.331617832183838, "identity": 0.6353833079338074, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8f08fdf6477fa0364a01234d30b4a94f7c62658fa5ce1df87a9483040ba340f8.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "8f1d7116e3c3762814fc0df10483444bff1a937c06d4e393f96f37caac419959", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c008c266eefe751a7d98bbd2dd8d5cef97b71a47e7b0215a1d042bb8ba940059", "generation_seconds_batch": 21.17920785106253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2799493074417114, "target_z": 1.617187499463806, "ce": 6.1894097328186035, "genuineness": 4.193545341491699, "identity": 0.2683894634246826, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8f1d7116e3c3762814fc0df10483444bff1a937c06d4e393f96f37caac419959.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "8f217a28815a6cf3d31330ee16bdeafd4328f0a326b8fa33301f3bcfb678ac37", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a066eab748e15782d80f2faf3597bbcba3c4afdafa36ca35c73cc74f274098e", "generation_seconds_batch": 17.18187458603643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3885064125061035, "target_z": 1.4921874908847026, "ce": 6.332377910614014, "genuineness": 4.281409740447998, "identity": 0.7002732753753662, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8f217a28815a6cf3d31330ee16bdeafd4328f0a326b8fa33301f3bcfb678ac37.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "8f4ead57139539b2f6ab111c86ea9422808998cc12222089c0530f68eb347547", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eeae94ba34bf14ce29219d8720737e8b4735e850ce23317f2728ed49832294e8", "generation_seconds_batch": 25.00598494906444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7168093919754028, "target_z": 2.265624966755974, "ce": 5.792276382446289, "genuineness": 3.8797430992126465, "identity": 0.5417078733444214, "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_VOLT_reg_low/8f4ead57139539b2f6ab111c86ea9422808998cc12222089c0530f68eb347547.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "8fa9b0eb36ba7d4d417e87d5eea7eba666521c2c048bb4c3a9e293d7d4e14d94", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca2184729d59b95e44c543fa82c0420ec5fdb40e60ca6332af4f80e79a054f5f", "generation_seconds_batch": 15.883588438038714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.197352409362793, "genuineness": 4.030367851257324, "identity": 0.4250287413597107, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/8fa9b0eb36ba7d4d417e87d5eea7eba666521c2c048bb4c3a9e293d7d4e14d94.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "902461ff59710fae498114340d433103abd8cd5f794c51042f350e68bc62b149", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d1c1115dd5968d58ae1af508512d1fbc9f658ce8899f5b77532d5499c92d2d8", "generation_seconds_batch": 22.70895272307098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.149696350097656, "genuineness": 4.068024158477783, "identity": 0.7683784365653992, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/902461ff59710fae498114340d433103abd8cd5f794c51042f350e68bc62b149.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "90cfd66ca728b96079ab206bc1230cce84823a87617bd71037f484d0da031eb4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "deaa0111d75ab030c2efe32ab5487de7cf3c4c42b5d478476ee7d9ea68178fe4", "generation_seconds_batch": 16.026584330946207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3885064125061035, "target_z": 1.4921874908847026, "ce": 6.268836498260498, "genuineness": 4.6705241203308105, "identity": 0.6694304943084717, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/90cfd66ca728b96079ab206bc1230cce84823a87617bd71037f484d0da031eb4.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "9144ba01a5517bae1c167c2cd9b310dfc624674a043c59029243b36b8137a1a9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28067b3f58881de0e52e6298be2551ca564ce31c4bfaf5ff957c3c0297c90585", "generation_seconds_batch": 22.720765540958382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.070269584655762, "genuineness": 4.432034492492676, "identity": 0.7608277797698975, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9144ba01a5517bae1c167c2cd9b310dfc624674a043c59029243b36b8137a1a9.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "9175f93ad7661f36e2f1f5d14ee58de6cb6307639b1db774a3ebec7f53ebc3e3", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a2002802cebd3ef7556eb964dc306c1836b9e14bc3f9c569f79ff7d98028f7d", "generation_seconds_batch": 15.823167231981643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.197352409362793, "genuineness": 4.532451152801514, "identity": 0.6755034327507019, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9175f93ad7661f36e2f1f5d14ee58de6cb6307639b1db774a3ebec7f53ebc3e3.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "91b941084b58557ffa9863a0790947482927c21ff8d852478ef301e0d3d623b0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f58c4cc5a9924d1020b8ede1cdac671db5556871d1b67af3f1ab8f99081e5704", "generation_seconds_batch": 22.947837801999412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.300607204437256, "genuineness": 4.432034492492676, "identity": 0.7651004791259766, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/91b941084b58557ffa9863a0790947482927c21ff8d852478ef301e0d3d623b0.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "922ce27e6b55be72d492aeb16c2b752272b7b22609b513546f315caca99433b4", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "223cad9b9b9b40a5363aa5dd81d82222be6243593c8f7035442686a697615207", "generation_seconds_batch": 14.632096879999153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4970635175704956, "target_z": 1.3671874823055992, "ce": 6.38003396987915, "genuineness": 4.406930446624756, "identity": 0.6669884324073792, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/922ce27e6b55be72d492aeb16c2b752272b7b22609b513546f315caca99433b4.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "92911dd29ac8e49d4324bc210af6b042b1a61c2653a8fa09f1280726d723e1e3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "16128b59380d6f2e0fd6e9f3201ffed5652032629a37713202a1dd7cd348f80e", "generation_seconds_batch": 24.14281178091187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.205295085906982, "genuineness": 4.770940780639648, "identity": 0.7139428853988647, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle a.", "wer": 0.05263157894736842, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/92911dd29ac8e49d4324bc210af6b042b1a61c2653a8fa09f1280726d723e1e3.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": "92b1bcadb18d93f3c0a0d9a8645d936b17a3b1a8ad73abc25dbbb2f2dc7ec60b", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "787edd9941ea25b3875bf5a182c19ca2fe5e44b53674b6b855e566b93cc3cfdc", "generation_seconds_batch": 21.720679554040544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.221180438995361, "genuineness": 4.005263805389404, "identity": 0.8111525774002075, "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_VOLT_reg_low/92b1bcadb18d93f3c0a0d9a8645d936b17a3b1a8ad73abc25dbbb2f2dc7ec60b.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "93d0b444d7e24225d40d01358411bb9ee5d025bbb8a237c76fa02dd2afed3871", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95ad299eb35069679a94ed61d9754f3702f6de924025c1018a42baee4bc133e7", "generation_seconds_batch": 22.301178419031203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.173524379730225, "genuineness": 4.15588903427124, "identity": 0.48882856965065, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/93d0b444d7e24225d40d01358411bb9ee5d025bbb8a237c76fa02dd2afed3871.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "93d10d8d3feb6dbd7a381fb10fbc88d668cc26d0bc68a9a595c74e08e38cad53", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "697834e37b12dcbfac76a4860366171912e98f68e41c3401eabae364bf642873", "generation_seconds_batch": 15.997215724084526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.117113709449768, "target_z": 1.8046874436996334, "ce": 6.395919322967529, "genuineness": 4.042920112609863, "identity": 0.5497473478317261, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/93d10d8d3feb6dbd7a381fb10fbc88d668cc26d0bc68a9a595c74e08e38cad53.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "93db48c62e168934cbdb2e5866641aa62e069c5e7b0aa323320f2b5893822621", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50fd78d75d53c1410500dcfd02cf2ea09ebf9c329c39217a8db4039173ac2700", "generation_seconds_batch": 17.23903857404366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.070269584655762, "genuineness": 3.9927117824554443, "identity": 0.4356963038444519, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/93db48c62e168934cbdb2e5866641aa62e069c5e7b0aa323320f2b5893822621.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "93fc854cf57e6f11e4d3f2350e63c2661822c2445870c9c865adafd0e8003930", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10f2f83b7dd8043da54c39acae28f1787dbda9cddcf8c9f13b6c2efd61d1f9eb", "generation_seconds_batch": 20.618647772003897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.395919322967529, "genuineness": 4.381826400756836, "identity": 0.39214885234832764, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/93fc854cf57e6f11e4d3f2350e63c2661822c2445870c9c865adafd0e8003930.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "942114c34364fa6b9814ebe27eae7fbd02792da4707c83d13bcc68e731e207cc", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c60b88fb76c4d14fdd2491305bc88e4351c4aa675439cdd6fd7af134d8cc568d", "generation_seconds_batch": 15.34919295005966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.316492557525635, "genuineness": 4.15588903427124, "identity": 0.6064543724060059, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/942114c34364fa6b9814ebe27eae7fbd02792da4707c83d13bcc68e731e207cc.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "9441559b73a5e1add10b531f890e740da723b651b872fcacdbced3964144c051", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2931292f7c24db77fba59d621f8d42d6eedb2097a9741275f0667d9b4e514186", "generation_seconds_batch": 20.677944466937333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 6.09409761428833, "genuineness": 4.005263805389404, "identity": 0.20781837403774261, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9441559b73a5e1add10b531f890e740da723b651b872fcacdbced3964144c051.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "94473d3bae46cd6968950fb7f5630962347a54c092265fc30bcf76d71dab47a7", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18433da131f5dcc2c8956c2113f0febbbc90197453ed56da311074bc0a0e7395", "generation_seconds_batch": 14.994901362922974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5445572137832642, "target_z": 1.3125000214477587, "ce": 6.165581703186035, "genuineness": 4.356721878051758, "identity": 0.6209489107131958, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/94473d3bae46cd6968950fb7f5630962347a54c092265fc30bcf76d71dab47a7.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "948d465f0b227f57006c963ac71209b6173bd9c8b4300ab40a18ddd88233b8e6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47d997629d83fecbf10376102f0c042e4bf26c4a154a4c278ba269a5542266d4", "generation_seconds_batch": 14.871567584923469, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.38003396987915, "genuineness": 4.381826400756836, "identity": 0.2938269376754761, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/948d465f0b227f57006c963ac71209b6173bd9c8b4300ab40a18ddd88233b8e6.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": "94b51ea9823373604af8d99886d2c720c2969c7e9f47f866df396069127228f2", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd8e9eb105a64e63b4ec456168af5f7ea869e5b75a0af735337c465f9d553de9", "generation_seconds_batch": 19.768817762029357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.324435234069824, "genuineness": 4.300237655639648, "identity": 0.6534465551376343, "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_VOLT_reg_low/94b51ea9823373604af8d99886d2c720c2969c7e9f47f866df396069127228f2.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "94d684eac3a24fd785f2c774a6d414471770fb7cab4ddc7b2830b50d2be81c38", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cce5b494815e8e484ad7dec9eb1d202bf523067341296c0db6dc5f8f37a7055", "generation_seconds_batch": 21.656389169045724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.165581703186035, "genuineness": 4.162164688110352, "identity": 0.8118812441825867, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/94d684eac3a24fd785f2c774a6d414471770fb7cab4ddc7b2830b50d2be81c38.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "95b4f39c71643ceca16f3f5b603d377736c4e77ddad7bddcf98258997c21c47d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0f109c8a97f5e41a30d5dfc81e209aa1f8980749b9936748228760dc9294fd5d", "generation_seconds_batch": 21.410446002963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6625308990478516, "target_z": 2.3281249024126978, "ce": 5.692992687225342, "genuineness": 4.432034492492676, "identity": 0.7574008107185364, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/95b4f39c71643ceca16f3f5b603d377736c4e77ddad7bddcf98258997c21c47d.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "968a2990aeb6ac3bb6c640ac533094971ff89898784c3571192c6a7b48bf11b5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e354afad75924fb460b46b1334a358cafc62f7a3aa9374b4bd1d295276fbb678", "generation_seconds_batch": 18.04255881602876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 5.8875885009765625, "genuineness": 3.8044304847717285, "identity": 0.6425179839134216, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/968a2990aeb6ac3bb6c640ac533094971ff89898784c3571192c6a7b48bf11b5.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "968a9f1fb9b5b4f01253a1610025f36b54712c76b8c2a95317bb938fca463334", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3d23cb47dc96f107b9e9db403b6dab17000c40949555bbb40841dd60830911a", "generation_seconds_batch": 20.946646210039034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.5137717127799988, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/968a9f1fb9b5b4f01253a1610025f36b54712c76b8c2a95317bb938fca463334.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "96a1f998516cc1859dd21c7cf5d4feb24bc46a0b19eb7368ad536db011d7e22f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "020257688ee52c26391013747cd28225d4165576dfe7435413c18aac6b90b5f3", "generation_seconds_batch": 21.45952530601062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.648961067199707, "target_z": 2.3437501265417766, "ce": 6.24500846862793, "genuineness": 3.9927117824554443, "identity": 0.2573316693305969, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/96a1f998516cc1859dd21c7cf5d4feb24bc46a0b19eb7368ad536db011d7e22f.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "96de2836b8258ba5ae3bcd64be5235868b637488de9be053ff54e6c2d4b2f8e8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "40299a8ceed36075c8e90ec8414a5207ede6b42b168951ff2d9f260087881ac2", "generation_seconds_batch": 22.148645624984056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9678477048873901, "target_z": 1.9765624383376936, "ce": 5.863759994506836, "genuineness": 4.413206577301025, "identity": 0.770209550857544, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/96de2836b8258ba5ae3bcd64be5235868b637488de9be053ff54e6c2d4b2f8e8.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "96eed20356f54d4d1ef3886f09dd8ff8c55ff6cb1db24546cd02fc0003200896", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a20d64692bbd3622e6875a08d544928f6ded58138b19d0de607f1308171874b4", "generation_seconds_batch": 16.94474713294767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.356205940246582, "genuineness": 4.13078498840332, "identity": 0.6433984041213989, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/96eed20356f54d4d1ef3886f09dd8ff8c55ff6cb1db24546cd02fc0003200896.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": "97921a799c4e8bf9c1cd60b11357cc91d07ad263ba54ca307924efa2fd17e560", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "27e1e6f37c9806f43e11c594c0b654fa9f865c209910639c94ba8555ccdd7d9c", "generation_seconds_batch": 14.844497876008973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 5.998785495758057, "genuineness": 4.2249250411987305, "identity": 0.814939022064209, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/97921a799c4e8bf9c1cd60b11357cc91d07ad263ba54ca307924efa2fd17e560.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "97ee55b58bef2a84c2fa31787f3dafed03001e8f08e468f4d3b98bf2c23417d3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3bc15843c0e28842a7e9a22645759366c7ae80607e9b8bd9f493c86f1a28d4c", "generation_seconds_batch": 21.9926674090093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.284721851348877, "genuineness": 3.9676074981689453, "identity": 0.2655961513519287, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/97ee55b58bef2a84c2fa31787f3dafed03001e8f08e468f4d3b98bf2c23417d3.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "98d0d8af68a218e1d45a48b9637ca4d35cf4218d631cad3e94edd156ad0c0bfb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b47bcffe2adfcbfe534ebdfcf243cea72865608cbb6a77fc312b06b57b3e4a0", "generation_seconds_batch": 23.10123184893746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.340320587158203, "genuineness": 4.532451152801514, "identity": 0.6476280689239502, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/98d0d8af68a218e1d45a48b9637ca4d35cf4218d631cad3e94edd156ad0c0bfb.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "98df8eb3ef9d4ef93b765d4e59634204419a681ed71a1d71bbf3bf4d40526ff5", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0cddda7339821f9050bf45944536b15c77c4520f294b2f3170dd4b749e297e0", "generation_seconds_batch": 16.500968012027442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.229123115539551, "genuineness": 4.362998008728027, "identity": 0.6609886884689331, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/98df8eb3ef9d4ef93b765d4e59634204419a681ed71a1d71bbf3bf4d40526ff5.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": "993ca0359509b049d32ca8ecb6c24a89a4eb20c2ca5325d25284930db2d5722b", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa64ee9d360b9e0cc77af269f1f3076670a7b7d9cd096bd3b9fab399883589a4", "generation_seconds_batch": 20.588579004979692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.475345611572266, "genuineness": 4.055472373962402, "identity": 0.3877919018268585, "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_VOLT_reg_low/993ca0359509b049d32ca8ecb6c24a89a4eb20c2ca5325d25284930db2d5722b.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "99a42feb049bd027c9a14e0348a27e55371d3c8138d65945862532bafc26263c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8711785e9262f016c68582b011a3fdd1edb066ca4fd4e39a81d31f280146ba2", "generation_seconds_batch": 15.33289022394456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.268836498260498, "genuineness": 4.3881025314331055, "identity": 0.2745434641838074, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/99a42feb049bd027c9a14e0348a27e55371d3c8138d65945862532bafc26263c.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "df79d8d9a1e957ce50cdf7f38ec058ecc1f6d8be45af12fb31349c3ea604852d", "generation_seconds_batch": 18.71632192307152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.476709008216858, "target_z": 1.3906250439679053, "ce": 6.2767791748046875, "genuineness": 4.557555198669434, "identity": 0.5600099563598633, "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_VOLT_reg_low/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.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": "9a289004acaf14570fba6ebfecdd244d6362c3ebfd60e0fa8f9410f1c1762e38", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4b718ad2afa9e9fdd6758905a552644cc177dcd1d80670738bf9664bdac5afe", "generation_seconds_batch": 14.939246786991134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.443574905395508, "genuineness": 4.13078498840332, "identity": 0.3642040193080902, "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_VOLT_reg_low/9a289004acaf14570fba6ebfecdd244d6362c3ebfd60e0fa8f9410f1c1762e38.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "9ac58174c2a779feea6d89afc125e0af198484c80d756a0e7a5af6a074c8c0d3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0876d26c097f777a9ba9c4a12419230ed5269640e46b69e3ebd6b93ecf6fda8e", "generation_seconds_batch": 16.147964627016336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.7961338758468628, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ac58174c2a779feea6d89afc125e0af198484c80d756a0e7a5af6a074c8c0d3.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "9ad2fe0e959ad85e2cb8cac0635eb49a3e6c55399cafc8227a098fe4186f510c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3cc8a2d2e69968faca5b2a2febe3aa6056f3f9a1d26146bd88d8f644a0f89130", "generation_seconds_batch": 15.029614907922223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.268836498260498, "genuineness": 3.9425034523010254, "identity": 0.49410009384155273, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ad2fe0e959ad85e2cb8cac0635eb49a3e6c55399cafc8227a098fe4186f510c.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "9ae08a2b5b09fee9ec8fe9aa44d30efb1981521f9701ce10b73e097744d9fc4b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a30557d459ff97d526cb51f47b5fc5f911b1c5111db46638f5d630e7e3e4c1d", "generation_seconds_batch": 20.908492472022772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6124054193496704, "target_z": 1.234374998927612, "ce": 6.006728172302246, "genuineness": 4.168440818786621, "identity": 0.5987619161605835, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ae08a2b5b09fee9ec8fe9aa44d30efb1981521f9701ce10b73e097744d9fc4b.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "9b1af611d19f6bf3fc6c0d937ecab5e04f3b14c00e56781adeee9b07e390d2cf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48c0e9268ca8f6df8f569b11aa48a0d35733aa22e4fa7070d502875d2122854d", "generation_seconds_batch": 21.474002231960185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.205295085906982, "genuineness": 4.444586753845215, "identity": 0.5640050172805786, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9b1af611d19f6bf3fc6c0d937ecab5e04f3b14c00e56781adeee9b07e390d2cf.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "9b68510bbf07cdc87edbd5aa059f10ef84a7ade9f3e46b331871c1c560451a90", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b2b8304aeaa42c6e298480f95fbfbac6a88f6673177c2b940234707b99392525", "generation_seconds_batch": 23.118953573983163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.676100492477417, "target_z": 2.312499952814931, "ce": 6.1894097328186035, "genuineness": 4.111956596374512, "identity": 0.7003051042556763, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen inleven.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9b68510bbf07cdc87edbd5aa059f10ef84a7ade9f3e46b331871c1c560451a90.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "9ba93a89a73e1044fca61f2457f2cc633de21a0a3378e467413c18fa236f1a43", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9182aa5fd3b4996d78334d3150527a9352c4d2d023c166eb5d541202e118663d", "generation_seconds_batch": 22.1537480809493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.1020402908325195, "genuineness": 4.18099308013916, "identity": 0.5430277585983276, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ba93a89a73e1044fca61f2457f2cc633de21a0a3378e467413c18fa236f1a43.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "9bc1483b05410932080cdb2e3844be11d14d0b90daf399355739be09dc7ddba9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "349a5c59eda85ae95d85211760bf81e1c8d0bb86ea8815053816561b62929b4e", "generation_seconds_batch": 21.63251548295375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.213237762451172, "genuineness": 4.105680465698242, "identity": 0.2720561623573303, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9bc1483b05410932080cdb2e3844be11d14d0b90daf399355739be09dc7ddba9.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "9bd634e25bfa0af34e27199433892cd7f4ec155b868747e8f11fae5e1def200f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d424fbba141edbfe108a5012633427cf1a46ba9b303196cb949da23e81aa1110", "generation_seconds_batch": 15.978407702990808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.693823218345642, "target_z": 1.1406250268096985, "ce": 6.24500846862793, "genuineness": 4.193545341491699, "identity": 0.24703548848628998, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9bd634e25bfa0af34e27199433892cd7f4ec155b868747e8f11fae5e1def200f.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": "9bf46ed83e2e31a78e17d8f9f74a2663a9672c097d86c00eaa568cdf8fb4a1e3", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fedf2b2b192c5e2a732173a0700d820e0ff169e3b0a347d6e181d7d1c0b1071", "generation_seconds_batch": 19.341351500013843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.459460258483887, "genuineness": 4.381826400756836, "identity": 0.5595512390136719, "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_VOLT_reg_low/9bf46ed83e2e31a78e17d8f9f74a2663a9672c097d86c00eaa568cdf8fb4a1e3.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "9bf502fe65264efbc350b5825c8811ae24f322ad9dea2c916faccc508dea1595", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06840ccae701e3a173075dbdaa459c0e78056c14d7e752815cde2bc06bae8151", "generation_seconds_batch": 21.410214703995734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.38797664642334, "genuineness": 4.406930446624756, "identity": 0.3594730496406555, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9bf502fe65264efbc350b5825c8811ae24f322ad9dea2c916faccc508dea1595.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "fb010de3ee3a73995ec167c874644d05f3e93afad9800d5c54e48249fb30018c", "generation_seconds_batch": 15.515985617996193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5309876203536987, "target_z": 1.3281249710455256, "ce": 6.221180438995361, "genuineness": 4.432034492492676, "identity": 0.7410755157470703, "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_VOLT_reg_low/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "9c21e150a8af1032aa459a790694a2f3cece77777fcb2aa2e18e2abd5121fb7b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 130, "seconds_generated": 10.4, "mp3_sha256": "72389aff9eb15a630998021b99fec3a79c6afe198d4ffb7f6d47a60ca3806b32", "generation_seconds_batch": 21.94206143799238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7982271909713745, "target_z": 2.17187499463806, "ce": 5.879645347595215, "genuineness": 4.231201171875, "identity": 0.7390969395637512, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9c21e150a8af1032aa459a790694a2f3cece77777fcb2aa2e18e2abd5121fb7b.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "9c4ab57c276099ad1084830aa1a2ebcc7fffbe72b488c2916b25ea77dd84a9e6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffdfdd1607c4971e30d8b97153a9a27c1a0aaa65a8c3794cb803bcdb527ca34b", "generation_seconds_batch": 17.989715995034203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.117925643920898, "genuineness": 4.507347106933594, "identity": 0.7451595067977905, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9c4ab57c276099ad1084830aa1a2ebcc7fffbe72b488c2916b25ea77dd84a9e6.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "66bf83e53e99c1986a533c0dde18ec8fadf99614b48772188dfecd00beae4fac", "generation_seconds_batch": 15.554405028931797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 5.919358730316162, "genuineness": 4.243753433227539, "identity": 0.670825719833374, "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_VOLT_reg_low/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "9cc062a82c544f1d08c68cd531ce4b2ba62eda71f6dc76b632222668db60081d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ab54b20f58b6cf6b2aa2ebf78cd48a9115909073a469f627d64f339a81f6b99", "generation_seconds_batch": 15.68025570397731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.213237762451172, "genuineness": 4.607763767242432, "identity": 0.7020536661148071, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9cc062a82c544f1d08c68cd531ce4b2ba62eda71f6dc76b632222668db60081d.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "9ce26cf2d55c01cdcd52f6efa0ca283d42016d37a118a14e36898dc84a646275", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a414ad8c6d409a1d09d9b29fed35890ec95e4b91669d4d2bc69b753f6052b864", "generation_seconds_batch": 22.172430134029128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.165581703186035, "genuineness": 4.2374773025512695, "identity": 0.7941526174545288, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ce26cf2d55c01cdcd52f6efa0ca283d42016d37a118a14e36898dc84a646275.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "9d2e0a665ead026ab3ecc3cf29c3d8abbb1182939de38bf21eda0116c549821a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4b96510f2e181257b50331dcd7b86ef67b9de9b67c0498d0e3e855a73e920b96", "generation_seconds_batch": 23.115749009069987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8396968841552734, "target_z": 0.9726562009382519, "ce": 6.149696350097656, "genuineness": 4.607763767242432, "identity": 0.8162928819656372, "hypothesis": "De deur 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_VOLT_reg_low/9d2e0a665ead026ab3ecc3cf29c3d8abbb1182939de38bf21eda0116c549821a.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "9db88e3a592a6b7caf308ddc65b0211b60a3d22136a9adae446a24040109b6e9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0ed1323760d45d4dc9e62060d148d26c2adbf4f4a9f38041b034a4d0dbdffac", "generation_seconds_batch": 19.87145583704114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 5.95112943649292, "genuineness": 4.306513786315918, "identity": 0.7836271524429321, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9db88e3a592a6b7caf308ddc65b0211b60a3d22136a9adae446a24040109b6e9.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "9e231c88f4d2309f7b805c24704d7cdab969f31544a508f7cbee1c09edf1352a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35bb5aa662dd70ae1a6fc8c5ac18b65c1b5e4d5d73d5efbc7084f696989cfbce", "generation_seconds_batch": 21.144257399020717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 5.982900142669678, "genuineness": 3.7793264389038086, "identity": 0.6637911796569824, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9e231c88f4d2309f7b805c24704d7cdab969f31544a508f7cbee1c09edf1352a.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "9e43bee6cb4474e3b3f730dc072ba21820f0a1a006e6ccd6a3ec3946ca762c5d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5416f4e420d7753ee339a925e26b9182adbe64fd0e6774faab9c9d3d7837bed1", "generation_seconds_batch": 20.561360616004094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.0146708488464355, "genuineness": 4.532451152801514, "identity": 0.7837392091751099, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9e43bee6cb4474e3b3f730dc072ba21820f0a1a006e6ccd6a3ec3946ca762c5d.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "9e68af5872b5193303919d1eeb3c3be50361e177bfcab168124e1ea195d1a372", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3ad5ce401ffc0280d6980816c1549b3467854afaa1de01d01fb17b001cc78677", "generation_seconds_batch": 22.692731917952187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9814172983169556, "target_z": 1.9609374887399267, "ce": 6.340320587158203, "genuineness": 4.18726921081543, "identity": 0.46268659830093384, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9e68af5872b5193303919d1eeb3c3be50361e177bfcab168124e1ea195d1a372.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "9e6ab2a2b31831bf7409204d8e9a74036bd48f6130e981e60f3fd03271e36b3a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3061eebb0b6a726fd4ba64cc8d9252160243d18d50cac09d6b6652353106d9e4", "generation_seconds_batch": 21.063399850041606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 5.824047088623047, "genuineness": 4.030367851257324, "identity": 0.6092410087585449, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9e6ab2a2b31831bf7409204d8e9a74036bd48f6130e981e60f3fd03271e36b3a.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": "9e8cfc97467e7ee3ffa4187e2df1dd8b7e68c8481bf1c668c2c63fa14b6f7073", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5aefcee27c4ac8cc3ee10a4189d0625c6a18f7089ba791575b8e8fc591de8e44", "generation_seconds_batch": 21.494886144064367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.41180419921875, "genuineness": 4.457139015197754, "identity": 0.8319475650787354, "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_VOLT_reg_low/9e8cfc97467e7ee3ffa4187e2df1dd8b7e68c8481bf1c668c2c63fa14b6f7073.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "9ed7a0307d1142bc733fc22cec0f0e2c754a0b1004805435f7e8dcbf874c25d0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f89990e362891a55ac546eae8b29446826335a9b2dbf37d60d83dc3490193b9", "generation_seconds_batch": 16.733373833005317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.284721851348877, "genuineness": 4.086852550506592, "identity": 0.6388541460037231, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ed7a0307d1142bc733fc22cec0f0e2c754a0b1004805435f7e8dcbf874c25d0.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "9efb3d99817187633e694ac761400b82894681504806bbb15363b2fd136b80d1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8a10743ef35d95467f377fc0a5192579f90d4e830a18eea06181e48fe9d91455", "generation_seconds_batch": 16.539025965961628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.372091293334961, "genuineness": 4.406930446624756, "identity": 0.6140788793563843, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9efb3d99817187633e694ac761400b82894681504806bbb15363b2fd136b80d1.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "9f7517c389c0e223306f6e5d8399e5a7d79a6d16e198eb27c8ff9eaab6cf8b18", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56ab9d39b990618969a8fc5047853bad319e0d5282a9fe5724d824a7aa9046c8", "generation_seconds_batch": 22.22923138702754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.157822608947754, "target_z": 1.7578124576406766, "ce": 6.133810997009277, "genuineness": 4.325342178344727, "identity": 0.7729364633560181, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9f7517c389c0e223306f6e5d8399e5a7d79a6d16e198eb27c8ff9eaab6cf8b18.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "9fe4ee16970c82fdb0e3f1b16315be795caf844f09a231f20e53dbe5005d0712", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08d334cd2786d63c29d1459cc85a9e7585ab64e631334df52b30c7b60f919844", "generation_seconds_batch": 20.887599274050444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.229123115539551, "genuineness": 4.293961524963379, "identity": 0.7017008066177368, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9fe4ee16970c82fdb0e3f1b16315be795caf844f09a231f20e53dbe5005d0712.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "9fe7495948f8b88085e45c9b2b67695b48681574ea0dbda3562538b3852b8dee", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14abf50a521bae953dca232a85035e8917ba39810e16609a3b423ea33e8717ff", "generation_seconds_batch": 21.016303991083987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.062326908111572, "genuineness": 4.118232727050781, "identity": 0.6498926877975464, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9fe7495948f8b88085e45c9b2b67695b48681574ea0dbda3562538b3852b8dee.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "9ffcceb06134bdfea7748601bf439297028e47ce676ed2a74bcee3c4beec6b7a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "883ff1cd77f217550e2f3979ba994ed04c82b204a756728e8e8cae5df5f77f80", "generation_seconds_batch": 24.072198606911115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.181467056274414, "genuineness": 4.030367851257324, "identity": 0.20139053463935852, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/9ffcceb06134bdfea7748601bf439297028e47ce676ed2a74bcee3c4beec6b7a.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "a009c85ef982fb347de8e4cbee637a7b165663a43775d90ea31238e9e338d406", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f345e9a925fc9a79db05d654529f00723d9e182d59b25a786796d8439868963", "generation_seconds_batch": 21.963544397964142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 5.871703147888184, "genuineness": 4.13078498840332, "identity": 0.7381134629249573, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a009c85ef982fb347de8e4cbee637a7b165663a43775d90ea31238e9e338d406.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "a052f355d18cedfae2e72be464e2138e5590490bae9d60b33373065cbbd77da0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "30c30c902e9e9242cf59e5d145d886cea8217d96803dbe017340ab8ece2f49e5", "generation_seconds_batch": 22.546399872051552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088608026504517, "target_z": 1.468750066488052, "ce": 6.348263263702393, "genuineness": 4.193545341491699, "identity": 0.7150636911392212, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a052f355d18cedfae2e72be464e2138e5590490bae9d60b33373065cbbd77da0.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "a06f2b0561e93e902360ad9913d80ca7a1ff7b53c28434e7e798b6a4bb6d7798", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "504c99481c7c42764df281fc138baa910fae5030f07b3af20b9921c0b912ac6f", "generation_seconds_batch": 20.948732709977776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.022613525390625, "genuineness": 3.8922953605651855, "identity": 0.7558549642562866, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a06f2b0561e93e902360ad9913d80ca7a1ff7b53c28434e7e798b6a4bb6d7798.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "a0847e867e9a3448108ec94e678048f10028f51d6e338a6d2fa753c45e249131", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5785636309852ea9963a74d08bce13ebc48ab985cce72e2efd9724f7be64f4af", "generation_seconds_batch": 21.063399850041606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 5.720792293548584, "genuineness": 3.9425034523010254, "identity": 0.7938877940177917, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a0847e867e9a3448108ec94e678048f10028f51d6e338a6d2fa753c45e249131.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "a08f52dd13235b2adfeee99cb73bba72a978cf2b5dbeba962a1a42c756fe250c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9adc8fe14667e9fab4be5152017535e6e7d618382faa27529e91b0250f2d647f", "generation_seconds_batch": 23.908192328060977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 5.990842819213867, "genuineness": 4.607763767242432, "identity": 0.4886331856250763, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a08f52dd13235b2adfeee99cb73bba72a978cf2b5dbeba962a1a42c756fe250c.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "a12a9dcecb6a21bd111433b4ab5d9469984ce664238ac0d054bb9210c606dc99", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f933d10feb2c67eb61f43a00f91c724bf3a574156bd1ea13952b62530500c581", "generation_seconds_batch": 14.586830905987881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.24500846862793, "genuineness": 4.293961524963379, "identity": 0.429945707321167, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a12a9dcecb6a21bd111433b4ab5d9469984ce664238ac0d054bb9210c606dc99.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "a1354880fb1b9913b63e25d263df9ee51d10501d62043b45bd5aa1ee1924dccd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72bdae5884692a968e358c83993ad27c37193288ff9f81fbe802393baf3ebe25", "generation_seconds_batch": 24.072198606911115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.125868320465088, "genuineness": 3.9801597595214844, "identity": 0.3413443863391876, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a1354880fb1b9913b63e25d263df9ee51d10501d62043b45bd5aa1ee1924dccd.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "a144e4e1c21737f08dae21805fdb672c4dafcb7753760baaee9e5bb592aa35d6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f341d2116acd6277a6be9820525a5aabf3fe6bee9a48f741369a968ca87b8d3", "generation_seconds_batch": 17.65645062807016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 6.23706579208374, "genuineness": 3.578493118286133, "identity": 0.3575824797153473, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a144e4e1c21737f08dae21805fdb672c4dafcb7753760baaee9e5bb592aa35d6.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "a14dff6294aeb5a1dc42c756d2c5db3db9da70df2f045743a5eb1e0e3ed2ee2b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f385bfff6384320891e55468f3c19b869970a7430ab6504282fb21a42d70a0e4", "generation_seconds_batch": 19.855579255963676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.149696350097656, "genuineness": 4.042920112609863, "identity": 0.31317317485809326, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a14dff6294aeb5a1dc42c756d2c5db3db9da70df2f045743a5eb1e0e3ed2ee2b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "a17c2d33a7e04a63773b03dcc36a87d0d6eb4a34d7dcc965b165bf0aadb4b955", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8afccc591b10a57f69fa273e256ec8a3d3730f5b6c1304e115ca63d0f3fd97ee", "generation_seconds_batch": 23.417324324022047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.6740294694900513, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a17c2d33a7e04a63773b03dcc36a87d0d6eb4a34d7dcc965b165bf0aadb4b955.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "a19243b9d16ea235659f45a4831d136faa631535825edc907843a50f8b4a60ab", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8793e26547360a34f22a74361129a341b6881cb56fbd4dbe287dc5547c19d8f5", "generation_seconds_batch": 15.149914038949646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.117925643920898, "genuineness": 4.381826400756836, "identity": 0.728488564491272, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a19243b9d16ea235659f45a4831d136faa631535825edc907843a50f8b4a60ab.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": "a1cf2cc2d8a63f871eb97d6d469f400403bebb30b0827588517b1f9dc79d1b62", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8555f3097b3c28ba9232c610f01e0e26ab850e894ca8becd8fec7c3acd8b7c7", "generation_seconds_batch": 19.30227281700354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8253664970397949, "target_z": 2.1406249581768706, "ce": 6.006728172302246, "genuineness": 4.243753433227539, "identity": 0.6719444394111633, "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_VOLT_reg_low/a1cf2cc2d8a63f871eb97d6d469f400403bebb30b0827588517b1f9dc79d1b62.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": "a28d64671d4fe74638f99af5124c48e12dbba834cd57271bc2ffd2abc8e4e108", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1eb085e2a27cc73513a83532c3b2d2ac57b1a7bfe1e3d37d5ec9fcdeaca43eab", "generation_seconds_batch": 17.022747148992494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.268836498260498, "genuineness": 4.557555198669434, "identity": 0.6326435208320618, "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_VOLT_reg_low/a28d64671d4fe74638f99af5124c48e12dbba834cd57271bc2ffd2abc8e4e108.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "a297afa6db1d099dd98c3d6d4b8315ad2c0fd96ed76b21263af6a2cfab38bace", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aea8dc67d8a0daa69df30ce1497585e1053f5a648ba1301e6a8b82bcb6dafe1c", "generation_seconds_batch": 15.44864875706844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.5269696712493896, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a297afa6db1d099dd98c3d6d4b8315ad2c0fd96ed76b21263af6a2cfab38bace.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "a38d39c7c200cd867d772b3f9a5b6b90079f1c20a10efa40526ba449bb20a65e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1744bbf8ec52508f33233106050e1d23b890a5edc7d0c8660120c6e4007993e7", "generation_seconds_batch": 16.877042258973233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6531143188476562, "target_z": 1.1875000128686553, "ce": 6.221180438995361, "genuineness": 4.482243061065674, "identity": 0.6436619758605957, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het auke 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_VOLT_reg_low/a38d39c7c200cd867d772b3f9a5b6b90079f1c20a10efa40526ba449bb20a65e.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": "a3c7fe6c459adc59ef5444f0e3f0e2be2eda823b5dc5f920cac43c6347e2d203", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6cd5f044aa8615d2a50aed576c4fec5c3292e5c2564095448f5e1027ea77ea34", "generation_seconds_batch": 25.238112117978744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 5.927301406860352, "genuineness": 4.331617832183838, "identity": 0.8573354482650757, "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_VOLT_reg_low/a3c7fe6c459adc59ef5444f0e3f0e2be2eda823b5dc5f920cac43c6347e2d203.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "a437823d691d30bbe3680ddd39454b9b1e08d6d6e276d1e2682a7f15d06549fb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "575c4caa19b06d60b576d9be69739f165245af28e83575340b54ab95544205e4", "generation_seconds_batch": 15.031020099995658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.221180438995361, "genuineness": 4.281409740447998, "identity": 0.5272141695022583, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a437823d691d30bbe3680ddd39454b9b1e08d6d6e276d1e2682a7f15d06549fb.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "a448dad7547f67c34443d27e5b32135286bd3b7db161213568aa9b3f7b119db0", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "513273bf015e91047c9e809e1545e2c47889fa52606875f86c5b8927c525fe56", "generation_seconds_batch": 15.628013657988049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.300607204437256, "genuineness": 4.21237325668335, "identity": 0.6611199975013733, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a448dad7547f67c34443d27e5b32135286bd3b7db161213568aa9b3f7b119db0.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "a491117d1153d7b7e63301576180b4da41558fc653e236acf3285d6706dd23af", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69fa4a7a1afaab059e04863c49018c620dea04364615eb98d21b75c64655e7cf", "generation_seconds_batch": 16.321099309949204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5174179077148438, "target_z": 1.3437500579089485, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.6396157741546631, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre. 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_VOLT_reg_low/a491117d1153d7b7e63301576180b4da41558fc653e236acf3285d6706dd23af.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": "a4bb8464b102606fc33df8473999174b51e39c97b417935886cf6bcc999f4fd1", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55b54663413493b28a1b44987dc53a75edcc46b5f859417b577c99855695aa40", "generation_seconds_batch": 16.210583418956958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.09409761428833, "genuineness": 4.406930446624756, "identity": 0.8676319122314453, "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_VOLT_reg_low/a4bb8464b102606fc33df8473999174b51e39c97b417935886cf6bcc999f4fd1.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "a50a945799e5d68af6dc276d0c4c68d022075660794c4aa7856f7fff3d91e0a7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95ef1cda25156aa3493ca251d3efb21c7e266ed7254edb2082a5f80e969a538b", "generation_seconds_batch": 21.17920785106253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.181467056274414, "genuineness": 3.9927117824554443, "identity": 0.33458757400512695, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a50a945799e5d68af6dc276d0c4c68d022075660794c4aa7856f7fff3d91e0a7.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "a55a5a16898066a0f1e34e3d55d07a119f805c023ff9341461aca0be0fe46deb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "664eb300ed5da6aece3710cbfd85c8f41d75edc5b2bfcd66a2fc37b88430fdb2", "generation_seconds_batch": 23.10123184893746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.205295085906982, "genuineness": 4.193545341491699, "identity": 0.5893015265464783, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a55a5a16898066a0f1e34e3d55d07a119f805c023ff9341461aca0be0fe46deb.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "a5735e4a2e8eb2a9d8498a61aad6f9e437c1a5c0547ed4fe4b6ca0510b07513c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e07d04a2bbb33b91601be4e6a013fd68d3a26d998a4f6caa8b81990e251ab105", "generation_seconds_batch": 16.40866323490627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6666839122772217, "target_z": 1.1718750632708883, "ce": 6.022613525390625, "genuineness": 4.20609712600708, "identity": 0.7562001943588257, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a5735e4a2e8eb2a9d8498a61aad6f9e437c1a5c0547ed4fe4b6ca0510b07513c.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "a5760f25857a5a403fbaf394175f5d66e11af5da8f8ef123e2cb63853bfc3481", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "786a2da18859ae637a0eb3c13f58fae34fc7b27d99beb1416583bf1b4e6ec452", "generation_seconds_batch": 22.47063395602163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.09409761428833, "genuineness": 4.105680465698242, "identity": 0.8499740362167358, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a5760f25857a5a403fbaf394175f5d66e11af5da8f8ef123e2cb63853bfc3481.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "a59d6c04667a482e79b3bf1a93001b812e0eb488fe2dc7e6ae8f2f70145b8c11", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92fc9dbf778675bb03e5197800ca56f5f0e1428552182ea5d4fd2984fdd1b74a", "generation_seconds_batch": 21.659675510018133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 5.863759994506836, "genuineness": 4.231201171875, "identity": 0.7049484252929688, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos la ventana.", "wer": 0.23529411764705882, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a59d6c04667a482e79b3bf1a93001b812e0eb488fe2dc7e6ae8f2f70145b8c11.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "a59f48dbbdeaf1894acafbb7c0c554667cac5350697693c1dd540f3bc02da6c7", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db379a545690ebbd67730071ee2028703368f6c79cfdfbe1b57c85dd634fbe0e", "generation_seconds_batch": 15.33752458402887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0628350973129272, "target_z": 1.867187516622013, "ce": 6.23706579208374, "genuineness": 4.444586753845215, "identity": 0.5076166391372681, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a59f48dbbdeaf1894acafbb7c0c554667cac5350697693c1dd540f3bc02da6c7.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "a5d00e98e2baae556f925c1484ed48cb2cbc85bc73827b58677d2b1eae79ea04", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53ea7d56ce6eb2f055bf84b339ea12d1296bf570156560b36b5e4872bf671acb", "generation_seconds_batch": 22.47063395602163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.260893821716309, "genuineness": 4.193545341491699, "identity": 0.8284539580345154, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a5d00e98e2baae556f925c1484ed48cb2cbc85bc73827b58677d2b1eae79ea04.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "a5ee08a79930e0789ecaa3153442807d8613b08aca345e43754f991e90504381", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb61c5ff8211da4d8a8a6ad39b3c84f8675714128cdd511dc651463cecd87678", "generation_seconds_batch": 16.833947304985486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7209625244140625, "target_z": 1.1093749903485086, "ce": 6.38003396987915, "genuineness": 4.607763767242432, "identity": 0.2880258560180664, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a5ee08a79930e0789ecaa3153442807d8613b08aca345e43754f991e90504381.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "a62a0df6f5298c03027ce3409aa6078d512b989a693ddc092d867bd1f82984e8", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61878795e66483173ec2453a3208ad77e3f4d2f1ca5412bedd9b2d0c5f97adee", "generation_seconds_batch": 16.44835966802202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.340320587158203, "genuineness": 4.13078498840332, "identity": 0.25680288672447205, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a62a0df6f5298c03027ce3409aa6078d512b989a693ddc092d867bd1f82984e8.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "a68a5a690788ca910c5566321e9ea9adb66578a0a5e414196585431591b8017b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f324bad4f5b45fec22ae28400b982ebf6ba460cae7328eed4fee059a14462483", "generation_seconds_batch": 21.62704359390773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0764046907424927, "target_z": 1.851562567024246, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.6179304122924805, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a68a5a690788ca910c5566321e9ea9adb66578a0a5e414196585431591b8017b.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "a703eb6cd4ccb7d48c7b40f3a4a0ccd436353b5c71507901f16ee89e31f5ac37", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8eccaefc07b8b1a00f196233dcb4df2b5dcd75f6517dfd8df0503c8766ee4061", "generation_seconds_batch": 23.088737585931085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.0305562019348145, "genuineness": 4.337893962860107, "identity": 0.656073808670044, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a703eb6cd4ccb7d48c7b40f3a4a0ccd436353b5c71507901f16ee89e31f5ac37.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "a79d3de649568d0bca2c0f22af78bb24bdcd8ea1f14e8c35653b88d4ba8afb9c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a64b6dce870de569eef1932a0dc007f1b1da99a551d5ded1b8ef18d0cc01419", "generation_seconds_batch": 16.573208494926803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.632759928703308, "target_z": 1.2109374372653057, "ce": 6.324435234069824, "genuineness": 4.444586753845215, "identity": 0.5874108672142029, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a79d3de649568d0bca2c0f22af78bb24bdcd8ea1f14e8c35653b88d4ba8afb9c.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "a8278a96be17afcc2c76b2032d7deffa825061bf338b9c6e812c0c3f71e856cd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a05b6406864c0c2b880252b3e98055594d13c2241a151b3872640b2c77b3e2c", "generation_seconds_batch": 22.301178419031203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.0146708488464355, "genuineness": 4.162164688110352, "identity": 0.6787940263748169, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a8278a96be17afcc2c76b2032d7deffa825061bf338b9c6e812c0c3f71e856cd.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "a84c6146449061ddff9a45e71a048c2dcb6c2464f44917db2b885eb61add8ac5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0dbf866b6e070553b94e6452c0d6956e003a7a3bfba4ce954b96d3ac30010851", "generation_seconds_batch": 24.05549571395386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.149696350097656, "genuineness": 4.570107460021973, "identity": 0.808314323425293, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a84c6146449061ddff9a45e71a048c2dcb6c2464f44917db2b885eb61add8ac5.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "a84fb218a734f1da73ef990d1d2e2ccfe871c74263e5254ddef665059f9b65c3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "006d8d00a2f5356c64dc5a79160dbe81d03578bfb208df1dc87190937fe3a821", "generation_seconds_batch": 20.592647108947858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.117925643920898, "genuineness": 4.042920112609863, "identity": 0.6548111438751221, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a84fb218a734f1da73ef990d1d2e2ccfe871c74263e5254ddef665059f9b65c3.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "a8738789acc36272ccb03360e022d792455cff62ea975a7323d94395a17150f8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "957270c19301944b0579aa4568bd0da09cdcd12c26a3039eb1f050483d5df56f", "generation_seconds_batch": 19.606482703937218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5716965198516846, "target_z": 1.2812499849865688, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.16205564141273499, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a8738789acc36272ccb03360e022d792455cff62ea975a7323d94395a17150f8.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "a882bc4ad3c91f7a5219e4a95958faf82d5ddb8934a8bccadb03abbb3802d939", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b6224c2faa699853593a7dec030324618304ca280d4ee98d1e35cbb9cc5c2ee3", "generation_seconds_batch": 23.118953573983163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 6.165581703186035, "genuineness": 4.406930446624756, "identity": 0.5741178393363953, "hypothesis": "Ik 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_VOLT_reg_low/a882bc4ad3c91f7a5219e4a95958faf82d5ddb8934a8bccadb03abbb3802d939.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "ef4b68cdbefcc3d3f94f0894e5c9e0f0fa25437f93575994ed194b3b0fd5fa0b", "generation_seconds_batch": 15.749888869933784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.1020402908325195, "genuineness": 4.319066047668457, "identity": 0.660391092300415, "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_VOLT_reg_low/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "a8be03ab5e36d1e9285e9702706f97bacf0f45522671b58f64d7708831bc2d1b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "41668ce30782353af93708e66513f9c62a157211d368783d6e47f0189a3d687f", "generation_seconds_batch": 22.37410696398001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4631394147872925, "target_z": 1.4062499935656725, "ce": 6.538887023925781, "genuineness": 4.532451152801514, "identity": 0.35362985730171204, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a8be03ab5e36d1e9285e9702706f97bacf0f45522671b58f64d7708831bc2d1b.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "a9471c712ccfed12a3f9b6acc4c762d5318fb5cdddfdf24bab4216ea448b25e5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7a06fc8b71331ace2e806e896efc2d9574ce97d951b4ef5fd5b732b7e75aae4f", "generation_seconds_batch": 21.360620917985216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.395919322967529, "genuineness": 4.243753433227539, "identity": 0.5376126766204834, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a9471c712ccfed12a3f9b6acc4c762d5318fb5cdddfdf24bab4216ea448b25e5.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "ed87b1550a4e9866f38700e09be3f6653f988e4923c8f95c5a253ff7f5b7fc4a", "generation_seconds_batch": 14.545881725032814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5649117231369019, "target_z": 1.2890624597854523, "ce": 6.252951145172119, "genuineness": 4.557555198669434, "identity": 0.8325854539871216, "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_VOLT_reg_low/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "a9e86ea8440847f1decf36caf0818d969cfa5a94ca6e12777b759196300f3dd0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ca8a468e35f847fab57b27e9907690037ce0ad3152a14e0f105d0921aa20940", "generation_seconds_batch": 20.561360616004094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 5.760505676269531, "genuineness": 4.570107460021973, "identity": 0.4582397937774658, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/a9e86ea8440847f1decf36caf0818d969cfa5a94ca6e12777b759196300f3dd0.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "aab42518052bc398d70077083bd15c2983b2afad0ee5190f9879b06cbdc7c931", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "271f84aec62e4d4173a61c6bda9c5a357803e74f9abc0ad89af71c770a2d7e54", "generation_seconds_batch": 16.257212248048745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.356205940246582, "genuineness": 4.231201171875, "identity": 0.6890993714332581, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/aab42518052bc398d70077083bd15c2983b2afad0ee5190f9879b06cbdc7c931.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "ab21bef70402bfe1d7e605eba8c354c5ca1cdfed9ff7f2fe066c6109273caab9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5e420e94c4af367fd543db1670d0af9bd2fbc59bcdc6667233d55ba411dcc6be", "generation_seconds_batch": 22.438938518054783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.395919322967529, "genuineness": 4.319066047668457, "identity": 0.5885812044143677, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ab21bef70402bfe1d7e605eba8c354c5ca1cdfed9ff7f2fe066c6109273caab9.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "ab333dc6558ce3ecfe209e34afeb661dba3d2f785047cf1e86805c6640efc828", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80721135b2954934fef584461fede8531d3a2931a8ae6360c46c37deff4b6f14", "generation_seconds_batch": 23.386858014040627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6124054193496704, "target_z": 1.234374998927612, "ce": 6.046441555023193, "genuineness": 4.394378662109375, "identity": 0.6304836273193359, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ab333dc6558ce3ecfe209e34afeb661dba3d2f785047cf1e86805c6640efc828.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "ab4c1d89ab77f280f1fbae40142bbdbfdb9ed24fbe9193fa17e4efa1757c1a08", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "302d5fed47bb0b96a60d2d516ba9e5e9f03e8eb6d3db9f73779f220b1a353030", "generation_seconds_batch": 22.95010663208086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5649117231369019, "target_z": 1.2890624597854523, "ce": 6.0305562019348145, "genuineness": 4.193545341491699, "identity": 0.6937273740768433, "hypothesis": "Die deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ab4c1d89ab77f280f1fbae40142bbdbfdb9ed24fbe9193fa17e4efa1757c1a08.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "ab730dbc591eb1edcf71dc7de9d9c83a9386e50a92029d816071e30852c49086", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cd9fa711882085ca2d9f0100ca09645aae0b59ea6487a75958186502fd3e19b", "generation_seconds_batch": 23.08426889893599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.324435234069824, "genuineness": 3.8420867919921875, "identity": 0.2504981458187103, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ab730dbc591eb1edcf71dc7de9d9c83a9386e50a92029d816071e30852c49086.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "5e163ee13858b53b36ab7ecaadeb74dfed9482a509c7c03fab5c9fbb1943dee9", "generation_seconds_batch": 13.139942476991564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4970635175704956, "target_z": 1.3671874823055992, "ce": 6.427689552307129, "genuineness": 4.168440818786621, "identity": 0.4674757719039917, "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_VOLT_reg_low/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "ab7b2fb750a9fd558b9c3bf070229cfbe3e9c8e1dc888f16cd24beac2e7d393d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f6bb7f1ab63a061ca603b3f0098f486569b2d428e64f631d52a63b6022f445b3", "generation_seconds_batch": 14.928994455025531, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.117925643920898, "genuineness": 4.1998209953308105, "identity": 0.7364522218704224, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ab7b2fb750a9fd558b9c3bf070229cfbe3e9c8e1dc888f16cd24beac2e7d393d.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "ab879bc87f809ec2e568b1060871adb70ea85df6b4480b4d0acb478f24a5072d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5f8654cc76f332050f5613af81e8f01c3152d5f97b2bc6a3759ff090059d093", "generation_seconds_batch": 16.733373833005317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.443574905395508, "genuineness": 4.325342178344727, "identity": 0.7088901996612549, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ab879bc87f809ec2e568b1060871adb70ea85df6b4480b4d0acb478f24a5072d.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "ac916f435be2e1dbc3f341fc45c66baba7a896af90ac013522ba8cb748cebda8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "48c3ab32acc755bcb9aaa2c829efbd0957efd8a588c770c675f5f2afd76513a8", "generation_seconds_batch": 25.13170381600503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0356957912445068, "target_z": 1.8984375530832027, "ce": 6.157639026641846, "genuineness": 4.093128204345703, "identity": 0.4746801257133484, "hypothesis": "Die Tür 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_VOLT_reg_low/ac916f435be2e1dbc3f341fc45c66baba7a896af90ac013522ba8cb748cebda8.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": "ace1b3828dcfd65b325a43667842af365bcf502d6173fdb9904af1085b1569b0", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f3c49d47fccb7fa7f4b38d67ea50e5ed3fd2f5965420bab085e07ac0376f790", "generation_seconds_batch": 22.997547816019505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 5.982900142669678, "genuineness": 4.413206577301025, "identity": 0.7842912673950195, "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_VOLT_reg_low/ace1b3828dcfd65b325a43667842af365bcf502d6173fdb9904af1085b1569b0.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "aceb2ddf77e33075c9b782790dcf1455a72a03eb8350f35120562720e0660a65", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "764ff97db02085c9918635d5c126bafd757d1a16514dbdc2687d9ed33de52a00", "generation_seconds_batch": 19.066799959982745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.648961067199707, "target_z": 2.3437501265417766, "ce": 6.062326908111572, "genuineness": 4.507347106933594, "identity": 0.8431270122528076, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/aceb2ddf77e33075c9b782790dcf1455a72a03eb8350f35120562720e0660a65.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "ad46adecf4c87e024d05387258ab3cd6a94b9f7b07bda4fef2ff27c15d476b6f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5fa25017551bc6be4f5dc819814f60732cf1fa424ffa5aca60f71d887c2ca6b", "generation_seconds_batch": 19.992995400098152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.5709168910980225, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ad46adecf4c87e024d05387258ab3cd6a94b9f7b07bda4fef2ff27c15d476b6f.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": "ad91253c676d2a2e5e72b01ee2cf7b0d32460fc44fab66f25708b9c5e7003a1d", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4768a6ce7ec17c0725c4825603408932400060d633a51d8f88b0e017887fb280", "generation_seconds_batch": 21.720679554040544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.125868320465088, "genuineness": 4.356721878051758, "identity": 0.7418209910392761, "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_VOLT_reg_low/ad91253c676d2a2e5e72b01ee2cf7b0d32460fc44fab66f25708b9c5e7003a1d.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "4396ef023b8063caa9a86b6bfd46ca0325c2308d385451b51036d5dc87f69c49", "generation_seconds_batch": 15.28965494595468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8023803234100342, "target_z": 1.0156250182305948, "ce": 6.09409761428833, "genuineness": 4.400654315948486, "identity": 0.7074831128120422, "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_VOLT_reg_low/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "adee89965671167c4128535d6a8ee993e8a8fa8d8f5920f882bd959c909731fc", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68d011c78f96e3c650307c231c3cae0d78aec54d9b0b57edae5afae72bcc621f", "generation_seconds_batch": 15.663863289984874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.6214677095413208, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/adee89965671167c4128535d6a8ee993e8a8fa8d8f5920f882bd959c909731fc.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "ae06775d287f4b38943671fbb60d50f6c12910a79e51167165fb02a0f1d54aff", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "293953bfa8bc07c242e01ab3c12d7be961e8fa49b33cf9ff8cba920a06985760", "generation_seconds_batch": 20.78523567505181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7168093919754028, "target_z": 2.265624966755974, "ce": 6.165581703186035, "genuineness": 4.21237325668335, "identity": 0.7326889038085938, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ae06775d287f4b38943671fbb60d50f6c12910a79e51167165fb02a0f1d54aff.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "bb49f213c1dd5247281009513851fe73af54a50def420ecc2bfab1becfcfdf69", "generation_seconds_batch": 16.957193318055943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.41180419921875, "genuineness": 4.105680465698242, "identity": 0.29496845602989197, "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_VOLT_reg_low/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "ae4997a48d6a837d06df83751aeb4b4c4fabb325f44c65de6da232fba5cc107d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef00a3aa6abf694f3f53df432e5db261e1e1430f7f5f0a3116dc37b266683258", "generation_seconds_batch": 16.691507424926385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.229123115539551, "genuineness": 4.419482707977295, "identity": 0.6188862323760986, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ae4997a48d6a837d06df83751aeb4b4c4fabb325f44c65de6da232fba5cc107d.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "ae4ca044d3b22e94c53d2960bdfd55e80fda301cabd533de2253ad7a22aa2242", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f166481eb9b44a53b3e378d45da76f84b8000a086c298fa016b73876834bd3a", "generation_seconds_batch": 16.49261487601325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.476709008216858, "target_z": 1.3906250439679053, "ce": 6.221180438995361, "genuineness": 4.6705241203308105, "identity": 0.7543359398841858, "hypothesis": "La porte est encore est 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/manual-comparison-v1/audio/vn_VOLT_reg_low/ae4ca044d3b22e94c53d2960bdfd55e80fda301cabd533de2253ad7a22aa2242.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "aebbd0b1b6311cf25a4f23898fd9c34e9411ad735af7fef1769550a40606b669", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "c1154dcabc8d1d2b1107d184b1e69fd70157c06448efbb9be3ee90e17ca97c05", "generation_seconds_batch": 21.618945914902724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.125868320465088, "genuineness": 3.8169827461242676, "identity": 0.5465361475944519, "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_VOLT_reg_low/aebbd0b1b6311cf25a4f23898fd9c34e9411ad735af7fef1769550a40606b669.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "aecb0f5904d8818636d71925ae84db74fa5363e08657df7e329d02396cfdaa9e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8243b1a202c14f3a4e3e31e3ef92f9a826cc8edda548490d107e8444b1adee93", "generation_seconds_batch": 22.24301067297347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.6461050510406494, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/aecb0f5904d8818636d71925ae84db74fa5363e08657df7e329d02396cfdaa9e.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "af3b313d69931f4e232ba97431ac06dcae0d563da5abbb185fbe9f7afb94bcf8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "293d2337ec9463435ec3d5050943336e71d54abb395f9522abb78981800e3cfc", "generation_seconds_batch": 21.411216643988155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.205295085906982, "genuineness": 4.118232727050781, "identity": 0.7441729307174683, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/af3b313d69931f4e232ba97431ac06dcae0d563da5abbb185fbe9f7afb94bcf8.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": "af449a5b4ee3e47d7e0052d4bac17b3fa22de9f3f18b93ca5a94a1ab7ebbe860", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b76d4b4ce34af39184da12b7584f15575558d2f1fe1203b584818d0b61866df2", "generation_seconds_batch": 22.548764117062092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.260893821716309, "genuineness": 4.645420074462891, "identity": 0.6358152627944946, "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_VOLT_reg_low/af449a5b4ee3e47d7e0052d4bac17b3fa22de9f3f18b93ca5a94a1ab7ebbe860.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": "af702ba4f04bb2509cf546a649a5a526dbf08b38a65074087664ad2073824e99", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b0f446571b107a36d4e9ac65f45aad092dde8817298d02ef3c3fe650858ec18", "generation_seconds_batch": 19.341351500013843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8117967844009399, "target_z": 2.1562500450402933, "ce": 6.149696350097656, "genuineness": 4.13078498840332, "identity": 0.7739834189414978, "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_VOLT_reg_low/af702ba4f04bb2509cf546a649a5a526dbf08b38a65074087664ad2073824e99.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "af7c3121f978d5c9069b900783a02331847b7ef40194fc6a7e8b83ed58878b29", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28693d73f15a047b8227d2d67d175e25589019a940d4696c3c84400b6af9117f", "generation_seconds_batch": 23.294789405888878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.395919322967529, "genuineness": 4.406930446624756, "identity": 0.35919806361198425, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/af7c3121f978d5c9069b900783a02331847b7ef40194fc6a7e8b83ed58878b29.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "afc235881a1284e96d60024a660945adedd3117cc5bde22cc4fbeb7e8ac8b742", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36495c6b332ccf987541f28740bdda30a304a1902223beeafe3b4935455951b8", "generation_seconds_batch": 21.545071269967593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0764046907424927, "target_z": 1.851562567024246, "ce": 6.205295085906982, "genuineness": 4.13078498840332, "identity": 0.6949637532234192, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/afc235881a1284e96d60024a660945adedd3117cc5bde22cc4fbeb7e8ac8b742.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "b07cf459c7e23816f787264e9a17fc31eeabdc24072d4a6e27ca2500ccc4d56c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1750f428ada138e06b6fba9d5640e8ce7c1f15a466c18257af8befe26e6da22", "generation_seconds_batch": 26.89300576900132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 6.181467056274414, "genuineness": 4.18099308013916, "identity": 0.776650071144104, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b07cf459c7e23816f787264e9a17fc31eeabdc24072d4a6e27ca2500ccc4d56c.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "b10c55be8ff03ef0c12c63fcec38fe332ffb8d2f3558560c82156dbac3ed4117", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "240045d7ec02591f217d8d5c068a22b5f3a326707bd7083f445ff2e389d52098", "generation_seconds_batch": 25.13170381600503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.5579040050506592, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b10c55be8ff03ef0c12c63fcec38fe332ffb8d2f3558560c82156dbac3ed4117.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "b18d81e1d046b83bbd3cdcf1907e2ba628465692cdd4eb833685ee85b299b541", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b1e48e5b2a570908ce646d09fe0dc95bb978fb603e79b674fbb6a696b918e25", "generation_seconds_batch": 16.138175346073695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.760505199432373, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b18d81e1d046b83bbd3cdcf1907e2ba628465692cdd4eb833685ee85b299b541.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": "b1fe6ca3a04ef97497a3dd9dd4519f6940d468eb7ce388ea45a0df310b04a595", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6580965e0e5c5cec03d64613aab19dbc10f1befc3a0d8e3dcd0226f26f4360af", "generation_seconds_batch": 18.068065503961407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.173524379730225, "genuineness": 3.8169827461242676, "identity": 0.5240681767463684, "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_VOLT_reg_low/b1fe6ca3a04ef97497a3dd9dd4519f6940d468eb7ce388ea45a0df310b04a595.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "b203ca54a141fb8accfbeedfafadfd7fdc310bc4db080020392a37ebf55fa8fa", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0f4a2c15566baf37dbbd53059084d471eb2e5e39752d7de99eef59f9cdfacaf", "generation_seconds_batch": 21.04820782900788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.332377910614014, "genuineness": 3.9801597595214844, "identity": 0.2318497896194458, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b203ca54a141fb8accfbeedfafadfd7fdc310bc4db080020392a37ebf55fa8fa.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": "b245f49daec2e5a04c40d20aba2ebedbf771b7154d20363c262d2d1ad08b863d", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "919a604f55491f1ac0de94c019a963792c633105973b81dce6f4593a3af90db7", "generation_seconds_batch": 15.074642130057327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.046441555023193, "genuineness": 4.218649387359619, "identity": 0.6482808589935303, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hacerlo.", "wer": 0.05, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b245f49daec2e5a04c40d20aba2ebedbf771b7154d20363c262d2d1ad08b863d.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "b26cef3fc2d7ac6574bafe7c200184a7bb48019f5ba2bd15e381fac9768d94f1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57adad147ea6fe0ff7b2abcc99fdad61d86337fb59ecf09c459193b4a75f7b51", "generation_seconds_batch": 23.668913187924773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6395447254180908, "target_z": 1.2031249624664222, "ce": 6.395919322967529, "genuineness": 4.582659721374512, "identity": 0.49627572298049927, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b26cef3fc2d7ac6574bafe7c200184a7bb48019f5ba2bd15e381fac9768d94f1.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "b29a2153c4f1f88bc875f54675cd471ae97dc1675a13b79d2d76975b3f41e9c7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a2434a2ea6338c8c44acb1811dba1b1af7d255c82d0db297a843bd647807387", "generation_seconds_batch": 21.62753762502689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 5.974957466125488, "genuineness": 3.9676074981689453, "identity": 0.48472291231155396, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b29a2153c4f1f88bc875f54675cd471ae97dc1675a13b79d2d76975b3f41e9c7.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "b2b7b873c7cc95c79ac89d0093cbf7b48099d029a39605f44fa7e99b6c8ee943", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00785f10fe1dfa1a6169abfa98e64ff0132be91fe1062d52d6b1b48888d9c752", "generation_seconds_batch": 22.24301067297347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.109982967376709, "genuineness": 4.243753433227539, "identity": 0.6647507548332214, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b2b7b873c7cc95c79ac89d0093cbf7b48099d029a39605f44fa7e99b6c8ee943.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "b3311b221f43a1a36f110400d55a2787b76d81205f317836c2a93e9777b43179", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6341eba67dc4bf8a61d0ee01c49809381ae89f346cff19eb4907d8fb23b4dea8", "generation_seconds_batch": 21.51017820101697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0764046907424927, "target_z": 1.851562567024246, "ce": 6.149696350097656, "genuineness": 4.432034492492676, "identity": 0.6108580827713013, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b3311b221f43a1a36f110400d55a2787b76d81205f317836c2a93e9777b43179.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "b378b0f5e5138959285680ab084ca999c59cf609e82acca3c563f104fa59d5ff", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df1a649ccce0bfab0fc6795ea52c482dd13cac1a06db4d2b5011ecbc44cc2bb2", "generation_seconds_batch": 15.291850434965454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.7493532299995422, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b378b0f5e5138959285680ab084ca999c59cf609e82acca3c563f104fa59d5ff.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "b3d3d5d3cedae064070c06a3a85d8b311baa95393b6c7e9131e9295fc2479199", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd1a2093296e3b46c6748909543337da8ebf95c5fa51ce51159c3c9a4e690877", "generation_seconds_batch": 20.607824043021537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.1020402908325195, "genuineness": 4.444586753845215, "identity": 0.5589969158172607, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b3d3d5d3cedae064070c06a3a85d8b311baa95393b6c7e9131e9295fc2479199.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "b3e017ea88c0b51a77dde6d878b93ff2721de87f8565dd23834fb60ab9ec5bc8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdac210a58e9f2899ca84c524c010d685427f33d761ae389d7830eb5a8bea879", "generation_seconds_batch": 22.6369284739485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 5.95112943649292, "genuineness": 4.055472373962402, "identity": 0.7471776008605957, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b3e017ea88c0b51a77dde6d878b93ff2721de87f8565dd23834fb60ab9ec5bc8.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "52e271cec14a6861cee0c34255251f1701687a08c9fe3122543e8c8220fae84e", "generation_seconds_batch": 16.465191354975104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.205295085906982, "genuineness": 4.137060642242432, "identity": 0.558972954750061, "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_VOLT_reg_low/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "b43fc143ffc0fddf3c3a3dfaaceb2c5230b7ec63287aa31c9058b04a64204124", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76fa2981f7c6bccea70c7c9ebd3912ddc6d928980b501aeda60b01a9b3d36012", "generation_seconds_batch": 21.410214703995734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088608026504517, "target_z": 1.468750066488052, "ce": 6.419747352600098, "genuineness": 4.287685871124268, "identity": 0.25312480330467224, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b43fc143ffc0fddf3c3a3dfaaceb2c5230b7ec63287aa31c9058b04a64204124.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": "b465393f79e18b9989a76b394ca786c7489f24626729fb5f2f0fe5962af057b6", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3012249615721d05d0df4e2275596fdf9580de6bc10e7066a97dd85e5c6265e", "generation_seconds_batch": 22.572549707023427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.435632705688477, "genuineness": 4.218649387359619, "identity": 0.20154868066310883, "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_VOLT_reg_low/b465393f79e18b9989a76b394ca786c7489f24626729fb5f2f0fe5962af057b6.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "b487db25a9a45e02187eb28971dc89dc866a9fcc8e895afc3af38f6f720e92f6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a040ae27c4107f30161ccb222312c560f1e6873c69b2259e23356c8c70f0e60", "generation_seconds_batch": 16.276662845979445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.6064497232437134, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b487db25a9a45e02187eb28971dc89dc866a9fcc8e895afc3af38f6f720e92f6.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "b50bf128340939d218e0ee5bbf033150747b96d6ce157f5e2c46d11d4599fd9c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0d82e1c7657c4415ae17d80cbf024dfbcf2d18914610ca84df512981f90b598", "generation_seconds_batch": 19.142784680007026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.673468828201294, "target_z": 1.1640624512063489, "ce": 6.062326908111572, "genuineness": 4.18099308013916, "identity": 0.7130956053733826, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b50bf128340939d218e0ee5bbf033150747b96d6ce157f5e2c46d11d4599fd9c.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "b53d69a189da03e70b12f60509d916fab8c75fc90e4da163678913fb559a7055", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "502e4cafa174f29babff7383c435d26352331859d2051e4c4c512a1e0f863f75", "generation_seconds_batch": 22.479444041033275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.149696350097656, "genuineness": 4.381826400756836, "identity": 0.6417308449745178, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b53d69a189da03e70b12f60509d916fab8c75fc90e4da163678913fb559a7055.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "b57739e83b247bc46b03290bd49cd5611887fcb1f81145dca4ed358d9c5f7834", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67e5dca6de98bb837f60146ed1118633a73a08ec9682704a07fbd8daa451ab6e", "generation_seconds_batch": 16.500968012027442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.252951145172119, "genuineness": 4.632867813110352, "identity": 0.5881363153457642, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b57739e83b247bc46b03290bd49cd5611887fcb1f81145dca4ed358d9c5f7834.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "b5ffcd0100d166fa35e0ba400d703db73127d10408224eda19f012b66081bcd0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c33348c92452e0857859d6eab5899a230ecff7044d26e8c0522e9cc18e01784", "generation_seconds_batch": 13.595037271035835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.09409761428833, "genuineness": 4.356721878051758, "identity": 0.727590799331665, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b5ffcd0100d166fa35e0ba400d703db73127d10408224eda19f012b66081bcd0.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "b65833aee97174a345e56fcfa1f6c371d7e2666282bf5774448eeebf5a3d0683", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89893563b19f8267c4c4585cd1e806994098650b5b0304cfb60929cb6f1e0af5", "generation_seconds_batch": 15.937258654972538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 5.998785495758057, "genuineness": 4.218649387359619, "identity": 0.5933924913406372, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b65833aee97174a345e56fcfa1f6c371d7e2666282bf5774448eeebf5a3d0683.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "b686b24c6466b84a67305f845efb9a828b0f48006b952f2ea6a89261e56cb44c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f4796581d07bf156c0197bd62a564e444418f3a2286b40fa3371ff8446ce8609", "generation_seconds_batch": 17.39317965891678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 5.990842819213867, "genuineness": 4.406930446624756, "identity": 0.7360365390777588, "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_VOLT_reg_low/b686b24c6466b84a67305f845efb9a828b0f48006b952f2ea6a89261e56cb44c.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "b6aa7ceea0a052dfde312ba1417362ebd892e01ef9f2adda27fafa1508af620b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a85c06e64bc1dcd31fb70f72c1fa96622d8054f20298292e736543ceec4ebf7", "generation_seconds_batch": 21.220728573040105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6056206226348877, "target_z": 1.2421874737264955, "ce": 6.348263263702393, "genuineness": 4.375550270080566, "identity": 0.4735313057899475, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b6aa7ceea0a052dfde312ba1417362ebd892e01ef9f2adda27fafa1508af620b.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "b6d4354777601f6598716f0d1e8519400a8c82603a583c2ca2949da0438e2318", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cbc72411c9611241cf8b1ebd7eca18839d01bf191fc65eea7dcb40b056959f9", "generation_seconds_batch": 24.015142335905693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.3641486167907715, "genuineness": 4.099404335021973, "identity": 0.7427887320518494, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b6d4354777601f6598716f0d1e8519400a8c82603a583c2ca2949da0438e2318.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "b6d5e00ecf15e7992e72ed787cd48534f8bf54004f2a16f6eb8a62929d30f099", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "718b308fc94b41d6dcac801a9e64099700d48e34b823fcb2c4a148e73ed1dfe7", "generation_seconds_batch": 23.369914517970756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.7333564758300781, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b6d5e00ecf15e7992e72ed787cd48534f8bf54004f2a16f6eb8a62929d30f099.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "b7011acf27e80cde48d9b5eee01d7ee5b94a70c9ebdc71980b88646be0ad59fa", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f93c8b02bc5e27a7bae3e9cba1d2d5c2a30158ba6e86bf98d581fccfc73d10be", "generation_seconds_batch": 15.904924610978924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0221261978149414, "target_z": 1.91406250268097, "ce": 6.213237762451172, "genuineness": 4.532451152801514, "identity": 0.788646936416626, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b7011acf27e80cde48d9b5eee01d7ee5b94a70c9ebdc71980b88646be0ad59fa.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "b71d3b17a81b674aa0da0cab697ae52112644459849e9a45cd9832fe1bac8593", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "169b74cef7d9914545df99f0e42b02e573c87245d06d675f0d90763c1b0b6d6e", "generation_seconds_batch": 21.46703628601972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.6185341477394104, "hypothesis": "La porte 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_VOLT_reg_low/b71d3b17a81b674aa0da0cab697ae52112644459849e9a45cd9832fe1bac8593.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "b73b60cd2f8182fb995e659a8867406632c27a083f7978252944e1dd1c16e8d5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d32b50a10f0cac8c738a056f5a267e357e63bca0d6b67e57c8f50901ea7e7842", "generation_seconds_batch": 21.0266902979929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4970635175704956, "target_z": 1.3671874823055992, "ce": 6.133810997009277, "genuineness": 4.281409740447998, "identity": 0.6448661684989929, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b73b60cd2f8182fb995e659a8867406632c27a083f7978252944e1dd1c16e8d5.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "b7a1eeb6d9aa9702d3150adb279272969d94c200f908464396ea661a157142db", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "959dc485f27a410d4a3262d3ad77e1acb63fe807f79f913a5bfe1556e4b0866f", "generation_seconds_batch": 20.851867430028506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.221180438995361, "genuineness": 4.243753433227539, "identity": 0.5444020628929138, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b7a1eeb6d9aa9702d3150adb279272969d94c200f908464396ea661a157142db.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "b7d0d2cee7e6c4ba97a093d2f87930d50c14e0e1ccb078fa01bbaa8e3f00c8c4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c3223d29be318be106af0e3936cec0fa574946cf2c75c3c68a06f85f97e4d91", "generation_seconds_batch": 22.753765064058825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.3641486167907715, "genuineness": 4.444586753845215, "identity": 0.21928924322128296, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b7d0d2cee7e6c4ba97a093d2f87930d50c14e0e1ccb078fa01bbaa8e3f00c8c4.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "b7d3daed7d66c2bd59ce39e03676434d5f59beae3e2f65a5877d62ea7d366527", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e090928b3362f488631b61260bb2943555dcfff5ba13546072bf8be9d12be157", "generation_seconds_batch": 16.787683447939344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.252951145172119, "genuineness": 3.8420867919921875, "identity": 0.4220958948135376, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b7d3daed7d66c2bd59ce39e03676434d5f59beae3e2f65a5877d62ea7d366527.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "b836b75a784b91de450e0754ad90a05ba935fe9397d77cb5a481b05bbaad0948", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9098edca8be279dcac9d23a0cdb39b2a6efaaea5a5b4d92bce428fbf7333bbf9", "generation_seconds_batch": 20.592647108947858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.157639026641846, "genuineness": 4.293961524963379, "identity": 0.5782939195632935, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b836b75a784b91de450e0754ad90a05ba935fe9397d77cb5a481b05bbaad0948.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "b837e2081df854859f115cd6fd4e553b6fd8b3b9c784310c06291fa32aba234d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf9e15d9dc9dbce316457d3521feb93568fb71d1b1f6da08df21b9449fd6d688", "generation_seconds_batch": 21.074272376019508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.205295085906982, "genuineness": 3.9801597595214844, "identity": 0.4543534517288208, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b837e2081df854859f115cd6fd4e553b6fd8b3b9c784310c06291fa32aba234d.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "b874b823d9b55830105a8592c75fa2d89576ea74682ff1c4b94a6cc86e7eddb6", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "873727020f1cf78a25cb195c9a8d0f89b69c1aec2b655352cf0d5453ce33c0e4", "generation_seconds_batch": 16.877042258973233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.673468828201294, "target_z": 1.1640624512063489, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.6709517240524292, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b874b823d9b55830105a8592c75fa2d89576ea74682ff1c4b94a6cc86e7eddb6.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "b8969843134c2bd57030558d43cbf9a20e09755ca85f9a935d55644195d20233", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8810c54952a6b60c486ead39e7a0b1b76cc7bb5355862a4f61baccc9ffb787b0", "generation_seconds_batch": 14.929830803070217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6395447254180908, "target_z": 1.2031249624664222, "ce": 6.197352409362793, "genuineness": 4.394378662109375, "identity": 0.4555387496948242, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b8969843134c2bd57030558d43cbf9a20e09755ca85f9a935d55644195d20233.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": "b8c8380e21d9bed918870a379b131d40e5d4330b5822f423dea390bf6372b728", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bddcd258ebd231073dd734382659e3014147ac07b93eb2ccf9cafcb5863a40a1", "generation_seconds_batch": 22.691784015041776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.6241962909698486, "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_VOLT_reg_low/b8c8380e21d9bed918870a379b131d40e5d4330b5822f423dea390bf6372b728.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "b8ca385c143161feb393a93cabecac57cbdfea52cac5c763680fc07441383a10", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f515dbdd4c48ce64417d6318150a74206653ac9610344c99e97a91cd4d45cf4", "generation_seconds_batch": 21.15319903299678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9542779922485352, "target_z": 1.9921875252011165, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.6576097011566162, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b8ca385c143161feb393a93cabecac57cbdfea52cac5c763680fc07441383a10.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "b8f79b251f1f61ea7a7b53a27d7bc930c1156bf7fdca1977a1011836861c3971", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "486bb6f2f0bd7b08b82e12b972edd967e9a60f6644c26abb1ffc3538aef5e677", "generation_seconds_batch": 15.801718350034207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8023803234100342, "target_z": 1.0156250182305948, "ce": 6.109982967376709, "genuineness": 4.770940780639648, "identity": 0.7456254959106445, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b8f79b251f1f61ea7a7b53a27d7bc930c1156bf7fdca1977a1011836861c3971.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "b91b6a038b931657b59380f0b50bc881c496eaa57d8f23f603e51474831cda57", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "533aba9c6656c7b186896bcad2190f1711d4070c8997d2a032b45c5bfdef1247", "generation_seconds_batch": 15.606603272026405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.372091293334961, "genuineness": 4.218649387359619, "identity": 0.7484015226364136, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b91b6a038b931657b59380f0b50bc881c496eaa57d8f23f603e51474831cda57.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "b9506912894e1824943c63635512fef52cbaf961b509eb27823be5af913c63c7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72033bf280b5a28776830c521f85b5054f76dd5cc04306dbc8f8db18a1c21b3b", "generation_seconds_batch": 17.246131401974708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.109982967376709, "genuineness": 4.093128204345703, "identity": 0.7605321407318115, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b9506912894e1824943c63635512fef52cbaf961b509eb27823be5af913c63c7.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "b9ae28f38d0ae932ce6018b7f8ca73d7774114399d12cb5bc28604bf447aa857", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b53b4c981ad0888bcf060c444c1066d56450de479c384e9e899c92ed3fea907d", "generation_seconds_batch": 21.085848371032625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.340320587158203, "genuineness": 4.507347106933594, "identity": 0.26690778136253357, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b9ae28f38d0ae932ce6018b7f8ca73d7774114399d12cb5bc28604bf447aa857.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "b9d8babdb98ec98ca922fe5a36667b442c28088abe48f6c85c517f18229c0128", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e24b22e879cb3a70594143fb8f7cfb37f7690ffb063bee7b515007aced122bde", "generation_seconds_batch": 19.82065737899393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.125868320465088, "genuineness": 3.9299514293670654, "identity": 0.3202636241912842, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/b9d8babdb98ec98ca922fe5a36667b442c28088abe48f6c85c517f18229c0128.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": "ba245740af28ee045664bc7bdee66f8c6b2d8a01e98c7af2835d77ca01ecd130", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cbe9c8f0ef53e2c435579f1f2ad52ae1b8050a0fe32506619d1b702ceb3da184", "generation_seconds_batch": 21.411520162946545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6218218803405762, "target_z": 2.3750000257373105, "ce": 6.221180438995361, "genuineness": 4.218649387359619, "identity": 0.5366262197494507, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weg.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ba245740af28ee045664bc7bdee66f8c6b2d8a01e98c7af2835d77ca01ecd130.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "ba466ffd8c0bcae7e2c0a62836f55b53b58f99b61c5994d60338e494df30954b", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de974959ffceab4b2f09d5674d50a8690eaf3d4b5e6d3ea714568cbd229335f8", "generation_seconds_batch": 15.710697405971587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.348263263702393, "genuineness": 4.256305694580078, "identity": 0.42680734395980835, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ba466ffd8c0bcae7e2c0a62836f55b53b58f99b61c5994d60338e494df30954b.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "ba66da513ee47ac2e5039d8acbf192f289b945b4a9bba42b039333a8bca54fc4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "230562f75ede6c1e0dee895c150d26731a697f62c476ebb4f429365e4ff55d26", "generation_seconds_batch": 20.821451934985816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.133810997009277, "genuineness": 4.20609712600708, "identity": 0.7401843667030334, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ba66da513ee47ac2e5039d8acbf192f289b945b4a9bba42b039333a8bca54fc4.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "41af9760ce1236aab2d892389a7dbc38542f90ca6c7515725b30e44fca9780ed", "generation_seconds_batch": 15.203751761000603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.173524379730225, "genuineness": 4.545003414154053, "identity": 0.6359573602676392, "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_VOLT_reg_low/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "bb349c050029c3796226eff9ef5e8e25f59917b36550f1311de97913870665ee", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1c9a5ae562c7eb2df49519046427c241accabd7b2a3ac543317bd5f4ded863d3", "generation_seconds_batch": 20.276484642061405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8253664970397949, "target_z": 2.1406249581768706, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.7362589836120605, "hypothesis": "Ik 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_VOLT_reg_low/bb349c050029c3796226eff9ef5e8e25f59917b36550f1311de97913870665ee.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "bb5363972151d69c51e4811f7fd856504862199aee62543d5d6ba0e08dc8a506", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0af88b303c8362a90c51bf4247be0a9bb43f8d971d629cbc59453b3c3b0efc35", "generation_seconds_batch": 21.022987902979366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.6573140621185303, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bb5363972151d69c51e4811f7fd856504862199aee62543d5d6ba0e08dc8a506.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "bb94b9d370c7e46ea8bc41699b7c98bec55b0918929b98d00bac59cad7421c02", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "beb56b4150d1f4e0146182b3e56500a1dc965d3cea5bab203a1517f3cf0fd1c3", "generation_seconds_batch": 17.17496317392215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3885064125061035, "target_z": 1.4921874908847026, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.6867812871932983, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bb94b9d370c7e46ea8bc41699b7c98bec55b0918929b98d00bac59cad7421c02.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "bc20e71d4417da39f1f4d157fff9686496e3c8a32fdc041daf2cfa39760e8dc4", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fecd2da96dbd84d9f0d72b2483b441b5693eadba64cbec4992f2daf09ac8bc2", "generation_seconds_batch": 16.707084671012126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.109982967376709, "genuineness": 3.785602569580078, "identity": 0.16230884194374084, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bc20e71d4417da39f1f4d157fff9686496e3c8a32fdc041daf2cfa39760e8dc4.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "bc31b18ebbdfba0d632a0660d2b8baceee552b6f8a36cb26047048203eef2717", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2585dca19f7f1f32c33d94459b1ce04b5ac1562465b672f946d70d4e223c66ed", "generation_seconds_batch": 22.11684586200863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.117925643920898, "genuineness": 4.369274139404297, "identity": 0.5186883211135864, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bc31b18ebbdfba0d632a0660d2b8baceee552b6f8a36cb26047048203eef2717.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": "bc8aff5144ab41203905b17340ca37fbf943d8517f7b728ea93b8a968fcad280", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "520287388b628d7954a7a588613e07b804cc3b6f36acaa33d0118c2b54006bd0", "generation_seconds_batch": 19.768817762029357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9814172983169556, "target_z": 1.9609374887399267, "ce": 6.141753673553467, "genuineness": 4.13078498840332, "identity": 0.541256844997406, "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_VOLT_reg_low/bc8aff5144ab41203905b17340ca37fbf943d8517f7b728ea93b8a968fcad280.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "bcdf96dfcc7cbc5446efd700ca417f40178bd6c529736abdfa910df0599e0e55", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c139a42df3cb7a44e058ab4151b333c9a89115ae50fe5f1beb1e61adc99b1b9", "generation_seconds_batch": 16.91272165288683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 5.967014789581299, "genuineness": 4.432034492492676, "identity": 0.732297420501709, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bcdf96dfcc7cbc5446efd700ca417f40178bd6c529736abdfa910df0599e0e55.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "bce1e61b655c4a82837b4af1b9ba3f0e1179ece83a2c0f835888d4edcf529a95", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91167e4c9e128a7943ff8d12d0dd72bea42fb88eac17d2b00236693c681c33c5", "generation_seconds_batch": 15.997215724084526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.23706579208374, "genuineness": 4.3127899169921875, "identity": 0.6236488819122314, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bce1e61b655c4a82837b4af1b9ba3f0e1179ece83a2c0f835888d4edcf529a95.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "bd2ad9fe2cc1fcfc6a76e7fde0c4950c6b26e24c82c5692589fb6a46c698747c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0400e088847309e0fe69121f25fd780ed81d2b1efc199a7b85fa00f2a975b009", "generation_seconds_batch": 18.03892247704789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4699242115020752, "target_z": 1.3984375187667888, "ce": 6.078212261199951, "genuineness": 4.808597087860107, "identity": 0.6896065473556519, "hypothesis": "J'ai trouvé la la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bd2ad9fe2cc1fcfc6a76e7fde0c4950c6b26e24c82c5692589fb6a46c698747c.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": "bd38e0c515d9d04b47fd9c8de77e935c6927165e7286247cb51346eb1473486a", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0af8f7d59efa95623663ec639df0dcf48d63a6c4f06f61229a41405b72db1bfd", "generation_seconds_batch": 16.93818897602614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.157639026641846, "genuineness": 4.381826400756836, "identity": 0.7719458937644958, "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_VOLT_reg_low/bd38e0c515d9d04b47fd9c8de77e935c6927165e7286247cb51346eb1473486a.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "bd676bc8da9fde1508ede41b0692da73be59d90610a553ee808817892a34fba8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7290b56598a3503fbb02262e4d29e0d4b039bad1026ee87bb5cc3eff6d326940", "generation_seconds_batch": 25.00598494906444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 5.720792293548584, "genuineness": 3.9801597595214844, "identity": 0.8309383392333984, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bd676bc8da9fde1508ede41b0692da73be59d90610a553ee808817892a34fba8.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "bd98788367cf3902f278f078d3870456f7abb8042fd5c0f065a022ed3b588556", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a0ab54c31be1c4151c980528e504cbbef12c428049bd48e2639d177c8432c94e", "generation_seconds_batch": 16.880071408930235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.6509417295455933, "hypothesis": "Ik 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_VOLT_reg_low/bd98788367cf3902f278f078d3870456f7abb8042fd5c0f065a022ed3b588556.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "be37d60d8fdb02203d410ccc8acceece30a227c86a0c8c5a9b5e8e62c906719c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbc0daaca950127adb8e277a45b869c920bcd5e2212378a90fa9a0161b5dcfb2", "generation_seconds_batch": 13.997606182005256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7439485788345337, "target_z": 2.23437506756044, "ce": 6.173524379730225, "genuineness": 4.15588903427124, "identity": 0.6381372213363647, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/be37d60d8fdb02203d410ccc8acceece30a227c86a0c8c5a9b5e8e62c906719c.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "be406f53ac9ada95958fb10ce4088304a76fd98f29d45df086d0ebb49c789e32", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5cbbfc3142e1102021e284d114d6a6d8a45632bd7ecf68265b07a5f95cbf74c", "generation_seconds_batch": 15.232622993993573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.3641486167907715, "genuineness": 4.532451152801514, "identity": 0.3858693242073059, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/be406f53ac9ada95958fb10ce4088304a76fd98f29d45df086d0ebb49c789e32.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "be583c40b6a78991c72d70933a38dc61c5590d436f041233af76bb2694f81306", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00c731fcd858aaf0f2a3fac55ae477dd56753a08b038742c774d84ec6be57481", "generation_seconds_batch": 21.345507033984177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 5.974957466125488, "genuineness": 4.20609712600708, "identity": 0.8089606165885925, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/be583c40b6a78991c72d70933a38dc61c5590d436f041233af76bb2694f81306.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "befb80974f290c333bf83424b870af1638bddf3756fb9c3e53f550a9ad4e5156", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e5eb2406a9bb4574f78b3ff1a648c5e2fa909be9f4a7fc6b5be48be0a839f7c5", "generation_seconds_batch": 21.393748603994027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 5.895530700683594, "genuineness": 4.2249250411987305, "identity": 0.35496827960014343, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/befb80974f290c333bf83424b870af1638bddf3756fb9c3e53f550a9ad4e5156.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "bf1735d0d563280c567e4a1bede7fb3e28606f78ac46361dc5ef6adb559a6055", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e51eff4465a8a9f30ee4377f93fe9d6181e2d214804f9066c79f65817c4b0d02", "generation_seconds_batch": 20.908492472022772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5377724170684814, "target_z": 1.3203124962466422, "ce": 6.213237762451172, "genuineness": 4.369274139404297, "identity": 0.7608786821365356, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bf1735d0d563280c567e4a1bede7fb3e28606f78ac46361dc5ef6adb559a6055.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "bf9a72116277df71f236a2f40a1407c8b9c62524c5eb0f38d23eaa50542e505d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efbc90dbecf8a0605df6d0c14a66f592b7a277bfa066bdf0aec65886f9c45f59", "generation_seconds_batch": 20.759771254030056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.7323559522628784, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bf9a72116277df71f236a2f40a1407c8b9c62524c5eb0f38d23eaa50542e505d.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "bfbbfbe72c4fe89aba6ae49a5158a014ae6094eabf9fdca3bca9a8e7b2cd2662", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5caa67088de32c4e0031947be3f931bbedb86f575537de72967f971fc4f2f8f5", "generation_seconds_batch": 24.05549571395386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.125868320465088, "genuineness": 4.325342178344727, "identity": 0.7154645323753357, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/bfbbfbe72c4fe89aba6ae49a5158a014ae6094eabf9fdca3bca9a8e7b2cd2662.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "c0384384f3808f260948ad32a240a656bb003323088f4d74f46dadf820d6973e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "74f7f8bc350775128af0f2ddc0ad79056434f75db53f49c348bbcaa1afb4f409", "generation_seconds_batch": 16.235591333010234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5174179077148438, "target_z": 1.3437500579089485, "ce": 6.372091293334961, "genuineness": 4.394378662109375, "identity": 0.4755402207374573, "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_VOLT_reg_low/c0384384f3808f260948ad32a240a656bb003323088f4d74f46dadf820d6973e.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "c149d08e1e78e1427df7e39c3f8cf1def804b616ac08e2b229a3ed54afddc5d5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66d56a8f1df328cf31672b0f380d66a037297bf2e3308970ec7ebae9e2f35b28", "generation_seconds_batch": 22.406015423941426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.229123115539551, "genuineness": 4.218649387359619, "identity": 0.741148054599762, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c149d08e1e78e1427df7e39c3f8cf1def804b616ac08e2b229a3ed54afddc5d5.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": "c16dfc72778a22c72d7ace57cca99dd6aeb62fc274ee6eebe765daf7a3973f98", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac65d74000e703d77befa7a466cca9d95a143518d6376a87757a020a6f2f7266", "generation_seconds_batch": 22.572549707023427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.374936819076538, "target_z": 1.5078124404824695, "ce": 6.427689552307129, "genuineness": 4.118232727050781, "identity": 0.33410143852233887, "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_VOLT_reg_low/c16dfc72778a22c72d7ace57cca99dd6aeb62fc274ee6eebe765daf7a3973f98.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "c262ba13ebaacd33cecb27ae5c0e64813481a133e01570c624477c0ee02b490b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cc00754960865511818ddb9c39e3c376de14005e6215a7f0e0effb63fc5585e", "generation_seconds_batch": 16.880071408930235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 5.863759994506836, "genuineness": 4.293961524963379, "identity": 0.7776294946670532, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c262ba13ebaacd33cecb27ae5c0e64813481a133e01570c624477c0ee02b490b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "c306d7827bddd30bdf862a87178ddfb4faf91f250ebb3bf5e13e2cbb27f8e04a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1395abaea8b025cbf3d5fac2c36feea1477d1b48717f8ac7b8b7fc045926901a", "generation_seconds_batch": 22.406015423941426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.181467056274414, "genuineness": 4.419482707977295, "identity": 0.7182129621505737, "hypothesis": "La porte est a d'un corps ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c306d7827bddd30bdf862a87178ddfb4faf91f250ebb3bf5e13e2cbb27f8e04a.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "c3135ab3cdd5883c45b9e2c4b6e1a1f8a887fd062953f24d9a1d98a931c5a2b5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12c9924f8ddf7e25efa5dd266253cb4485ac68c579fb760a2a508b2b54329428", "generation_seconds_batch": 15.904924610978924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.8552526235580444, "hypothesis": "J'ai trouvé la petite boîte, elle était près, de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c3135ab3cdd5883c45b9e2c4b6e1a1f8a887fd062953f24d9a1d98a931c5a2b5.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "c393ca17515ab2ee704d2c6b3b4091ff41d77926c584c68dfe312484f715d525", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba86bc2d34a625e41f479875fb270c32b3ad28c16791f83e4b3b2bab803613c0", "generation_seconds_batch": 15.854795605991967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7888107299804688, "target_z": 1.031249967828362, "ce": 5.998785495758057, "genuineness": 4.375550270080566, "identity": 0.7022457122802734, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c393ca17515ab2ee704d2c6b3b4091ff41d77926c584c68dfe312484f715d525.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "68f16f3a3971ea2f9f35b1fabdca8cde54bfc83c21696d2456e701f86e6ce89b", "generation_seconds_batch": 15.554405028931797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4631394147872925, "target_z": 1.4062499935656725, "ce": 6.086154937744141, "genuineness": 4.595211505889893, "identity": 0.7954391241073608, "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_VOLT_reg_low/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "c431ea8dd027857c7866e0d39d00d0550c67443948ae369daecdcb3714f03456", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9219d6e29352722cf453440211efa90ac3bdac635af7683e16c24f8fff0411fb", "generation_seconds_batch": 22.961514586000703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 6.165581703186035, "genuineness": 4.268857479095459, "identity": 0.8230146169662476, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c431ea8dd027857c7866e0d39d00d0550c67443948ae369daecdcb3714f03456.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": "c48ab466bf0b9c671df9868fc2f1d941a5558f04d022a37759acb0749197d2ec", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7251db09c71a7f2a4a9dd226f9e99b5dadf7c3709e76845d87a087a6e5b98dda", "generation_seconds_batch": 16.210583418956958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.221180438995361, "genuineness": 4.494794845581055, "identity": 0.7970873117446899, "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_VOLT_reg_low/c48ab466bf0b9c671df9868fc2f1d941a5558f04d022a37759acb0749197d2ec.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "c49132cefbab26b7dfe1ef3fc90816607b0b40dabc302264c6b5c35ab0a6f62a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84c6b912ffa6b686e96c41728cbe06281902df369cf63445459f889ce1137ece", "generation_seconds_batch": 18.937331393011846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.173524379730225, "genuineness": 4.3504462242126465, "identity": 0.6119391322135925, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c49132cefbab26b7dfe1ef3fc90816607b0b40dabc302264c6b5c35ab0a6f62a.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "de5a37e30e16b6f2a33ac921ab9654bf548ae9eb404090e01e1c7987a47deaca", "generation_seconds_batch": 15.515985617996193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6124054193496704, "target_z": 1.234374998927612, "ce": 6.197352409362793, "genuineness": 4.532451152801514, "identity": 0.6016448736190796, "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_VOLT_reg_low/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.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": "c497ae2ca6d7a50b40b82fa995e0bc5e5f9cdd90aff36833962173350d12ca00", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I wrote the number on the envelope.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e655554d7a0c80cdbe1b641bbc82a27be6d1cec0c500943580292536db70e838", "generation_seconds_batch": 14.939246786991134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.427689552307129, "genuineness": 4.15588903427124, "identity": 0.33420121669769287, "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_VOLT_reg_low/c497ae2ca6d7a50b40b82fa995e0bc5e5f9cdd90aff36833962173350d12ca00.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "c4c75bcf2d923ea6f9ba5c2895c43f3cb0161bc83e9e12d9327438167a1312f1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a3aa29f35c82c1f7dadc2cbf8375915373fae402d4ea6143599d0ec95adf3c2", "generation_seconds_batch": 15.883588438038714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.197352409362793, "genuineness": 3.9173994064331055, "identity": 0.24518680572509766, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c4c75bcf2d923ea6f9ba5c2895c43f3cb0161bc83e9e12d9327438167a1312f1.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "c4c9bc0e2d5879de8306a9cd61138c5dcfe11203a32720b421c6d2361626caaf", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08da6ca65431b76a4bd7e453da6e292f4e87344569f3385b59039fede08b68bc", "generation_seconds_batch": 22.971873180940747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 5.927301406860352, "genuineness": 4.118232727050781, "identity": 0.5697845816612244, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c4c9bc0e2d5879de8306a9cd61138c5dcfe11203a32720b421c6d2361626caaf.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "c4cf79fc0055681ff7246f2b764e4e8624b73bd8067d3e118fd3bef2ab7e1be7", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adfc3226597ff63c6ba533f9b81304b7b0e261bff973bf667489ce7f380a825f", "generation_seconds_batch": 15.735956192016602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.6260244846343994, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble what she 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_VOLT_reg_low/c4cf79fc0055681ff7246f2b764e4e8624b73bd8067d3e118fd3bef2ab7e1be7.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": "c526c91a581543320c81960027a14982cb54e8649123e0b3ddfef9b04a3e88b7", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ceee932b2b78779abc45153dbf23cbaf98fa4933576900819d9dd9ee60b4985", "generation_seconds_batch": 16.93818897602614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6463295221328735, "target_z": 1.1953124876675387, "ce": 6.284721851348877, "genuineness": 4.507347106933594, "identity": 0.7179436683654785, "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_VOLT_reg_low/c526c91a581543320c81960027a14982cb54e8649123e0b3ddfef9b04a3e88b7.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "c5ed05640641b5d151fe28c5a49bd9b3b0042759bb4723ded34e9d224f51efa2", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b1197fbc86231bfe6d423ce0d318fee370d40388956f8902390f12c02e82c1a7", "generation_seconds_batch": 23.42620371596422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 5.95112943649292, "genuineness": 4.055472373962402, "identity": 0.7821061611175537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de aquí.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c5ed05640641b5d151fe28c5a49bd9b3b0042759bb4723ded34e9d224f51efa2.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "c645b2ea37aea19af6d0360bc5c08fc0f3c16d5778b056592fbb138224d9b80b", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26fbf4907d49939c564e89741583bd6a823e349db5221ea08e79df725b9bcb22", "generation_seconds_batch": 16.277106542955153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.205295085906982, "genuineness": 4.15588903427124, "identity": 0.5750426054000854, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c645b2ea37aea19af6d0360bc5c08fc0f3c16d5778b056592fbb138224d9b80b.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "7c1ab83ef847298e0e9486045a34ee0d7efbf284bef377b4ee760c2614726884", "generation_seconds_batch": 15.28965494595468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.693823218345642, "target_z": 1.1406250268096985, "ce": 6.1894097328186035, "genuineness": 4.281409740447998, "identity": 0.6588940620422363, "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_VOLT_reg_low/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "c825af1ae52578d875ed8767785ce6edf2495cf08af01fcf128a7852958d554b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95870f1bcb41a61530d5e3f51859aecb65a755a37d5a77f74fb22baf9da97eda", "generation_seconds_batch": 21.360620917985216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.507116317749023, "genuineness": 4.256305694580078, "identity": 0.6269431114196777, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c825af1ae52578d875ed8767785ce6edf2495cf08af01fcf128a7852958d554b.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "15ccbe6340746b5853e0f7b63d87732541e8ad10f7f9a23a1c58a8033320bfce", "generation_seconds_batch": 15.863717218046077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.252810001373291, "target_z": 1.6484375359249959, "ce": 6.483288764953613, "genuineness": 4.419482707977295, "identity": 0.5093901753425598, "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_VOLT_reg_low/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "c920b9eeb57792435be7f1d4e9d2d82a3cd60cd5586d6e86f972c3aed84185e5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d65c4ae05626986dba2ed0ea4b33963307f28c95daa51b7235dcfa26ef850d78", "generation_seconds_batch": 20.113493154989555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.8168286681175232, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c920b9eeb57792435be7f1d4e9d2d82a3cd60cd5586d6e86f972c3aed84185e5.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "c920c3c75905eaff6bcdb7465b75b3cf8769b4a223d405babc538da895dc7d91", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "634406a33471c31cec9233bbb373fdb62dedc2ca09f7c784da4df97784bf4b61", "generation_seconds_batch": 23.0907962230267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.3314565420150757, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c920c3c75905eaff6bcdb7465b75b3cf8769b4a223d405babc538da895dc7d91.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "c99418ecb53fe301231515d2e9d770942a6d5729b0c6dae562a328e49d727188", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df7461b13af59a4f3d4e4ba33423a09201c474aaf534873f590186d8100dd16e", "generation_seconds_batch": 19.972900951979682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3206582069396973, "target_z": 1.5703125134048492, "ce": 6.3641486167907715, "genuineness": 4.375550270080566, "identity": 0.3220442533493042, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c99418ecb53fe301231515d2e9d770942a6d5729b0c6dae562a328e49d727188.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "c9e51c044f77c6c50e6076e48c2010bf1c02f057fde985d19c6cdad19a5b7173", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "c7a8e0440d91dbbec99ce17b4606bd8872690aacde917bb208e259da223d02b8", "generation_seconds_batch": 16.56943252508063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 5.7406487464904785, "genuineness": 3.591045379638672, "identity": 0.5623518824577332, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/c9e51c044f77c6c50e6076e48c2010bf1c02f057fde985d19c6cdad19a5b7173.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "ca58b230093f24b3ea8deb3da6a88e2edd7ecc12fa1af19a281e9c6b8824188d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "402e7b89f90213848caa4d2993e4cce0f1af1b8a419c3fc78de712d457857d9e", "generation_seconds_batch": 19.63305997592397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.2767791748046875, "genuineness": 3.9927117824554443, "identity": 0.3634394705295563, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ca58b230093f24b3ea8deb3da6a88e2edd7ecc12fa1af19a281e9c6b8824188d.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "cadb09f51f2ab2dc246c3fa6443eba70c589dd435b84cf9b98ee67856f805522", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2643b93d2495d721dda8efb5ff611b4f858fa60e5ddf4e55e69ea69f8c781a9", "generation_seconds_batch": 13.595037271035835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4292153120040894, "target_z": 1.4453125048257458, "ce": 6.181467056274414, "genuineness": 4.287685871124268, "identity": 0.6534323692321777, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cadb09f51f2ab2dc246c3fa6443eba70c589dd435b84cf9b98ee67856f805522.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "cadc27f9561bd537fe69f41ed92b44aa89935b1757a2f227ad59a853b686439a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5c20acca92b1fc24a4edb7fe8c78a6e7607f7fc35c8901625fd0178d49c9425", "generation_seconds_batch": 16.77278904907871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.63889479637146, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cadc27f9561bd537fe69f41ed92b44aa89935b1757a2f227ad59a853b686439a.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "cb0404c83343fb615c464cdf47bdd377afc05408ff7018a0533850124c7f710d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "be315b973d02163e4fc5d2af03b2cd4ed973f01dc75a91046362fb5027b54223", "generation_seconds_batch": 20.887599274050444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 6.260893821716309, "genuineness": 4.080576419830322, "identity": 0.6320943832397461, "hypothesis": "Ik 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_VOLT_reg_low/cb0404c83343fb615c464cdf47bdd377afc05408ff7018a0533850124c7f710d.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "cb0827443e69804321e61aeb22758dd5df3c2c726e31ca0d05ed06bed01ba34b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4db82784ae3409c052896799736951cf947a35c2d82f646db7df336530b09ca5", "generation_seconds_batch": 21.046147524029948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.144252896308899, "target_z": 1.7734375445040993, "ce": 6.372091293334961, "genuineness": 4.432034492492676, "identity": 0.2794869542121887, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cb0827443e69804321e61aeb22758dd5df3c2c726e31ca0d05ed06bed01ba34b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "cb1c594558184af7a4bc16e938ff626a01dbdc27c045508e977d54997b0e59b9", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "230cf313dda1b191afd70928e2623ca9459a708a62dfd89bc66e82dd890df4a3", "generation_seconds_batch": 15.761161356931552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.062326908111572, "genuineness": 4.137060642242432, "identity": 0.7176695466041565, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cb1c594558184af7a4bc16e938ff626a01dbdc27c045508e977d54997b0e59b9.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "cba9146efc0913b5cac95acc0b3a0fbef747e8dd2b278817f4e592d97569a9e7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65056d29f2d34458c66c7d55e244bd58a97757195ae60f72947100c9878bb94f", "generation_seconds_batch": 18.92224186391104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 5.911416053771973, "genuineness": 4.557555198669434, "identity": 0.7264691591262817, "hypothesis": "He encontrado. La cajita estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cba9146efc0913b5cac95acc0b3a0fbef747e8dd2b278817f4e592d97569a9e7.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "cc9f74f8f022c7fc193fa8ead5ffa7b842d42f274e39a8c4e393ae91bbc6c935", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ede1c04e02964d79c8bb9567147b919358d3dfc411e0213c53d26dcc3c0b0dba", "generation_seconds_batch": 21.926998381968588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7439485788345337, "target_z": 2.23437506756044, "ce": 5.935244083404541, "genuineness": 4.519899368286133, "identity": 0.8264155983924866, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cc9f74f8f022c7fc193fa8ead5ffa7b842d42f274e39a8c4e393ae91bbc6c935.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "cd64968069902ccd7192c639418528670a6176260ed571320a586fefd7123128", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad921ab16eae4c293fa55d502fa9254c38092cb91ce833cf0b3784efa378aca2", "generation_seconds_batch": 23.21826102794148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.45800745487213135, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cd64968069902ccd7192c639418528670a6176260ed571320a586fefd7123128.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": "cd9417d7bfaaf4c9eb39b90fe829dd381452cc047777c14131a94331ffc48b20", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The road turns left behind the bridge.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfdfb24214e46fea043b30922c5aab0a65bd6a36a9a3d72e3dc0f4bb7946cd45", "generation_seconds_batch": 15.487331610987894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 6.4515180587768555, "genuineness": 4.2374773025512695, "identity": 0.4738225042819977, "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_VOLT_reg_low/cd9417d7bfaaf4c9eb39b90fe829dd381452cc047777c14131a94331ffc48b20.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "ce23c0c9f813a5f09a066b0ae88341d483ee33249662fcd2d6e8b155e40abab7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d10b68e9138b3c996e64f45ce626f5d7657fd722d01ae8292626554c9aedc497", "generation_seconds_batch": 22.75454349303618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.078212261199951, "genuineness": 4.20609712600708, "identity": 0.7578418254852295, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble what she 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_VOLT_reg_low/ce23c0c9f813a5f09a066b0ae88341d483ee33249662fcd2d6e8b155e40abab7.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "ce4e01c2933ada2c97bc9f3ef1ffbfc2097f1214d381b43089e4f1ff02eb16a0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ff40a39e0991cbff761f1560cc9f7c9c5ab3f8b7bcec98cf23c9420134016f3", "generation_seconds_batch": 24.195031918003224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.006728172302246, "genuineness": 4.061748504638672, "identity": 0.7709561586380005, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ce4e01c2933ada2c97bc9f3ef1ffbfc2097f1214d381b43089e4f1ff02eb16a0.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "ce7a3a281609fc87f9d1bbafcec2c157a890b21414da2648b3f764a94d652707", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e7e6dcff9f46f52b4c0e351c93dc90890441a7c44697c56921b48f1b297fbfc", "generation_seconds_batch": 15.663863289984874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.427689552307129, "genuineness": 4.20609712600708, "identity": 0.45012882351875305, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ce7a3a281609fc87f9d1bbafcec2c157a890b21414da2648b3f764a94d652707.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "cec0666d25b150c781076e145348e70a5adad9c6d740518e0149241d441014f1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ae5c8341e7c86af71dbdf5b12bcd944d1163447ba3a316e8bd219f62b8cbf0e", "generation_seconds_batch": 16.143575005000457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 5.990842819213867, "genuineness": 4.231201171875, "identity": 0.724189817905426, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cec0666d25b150c781076e145348e70a5adad9c6d740518e0149241d441014f1.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "cf317b98f3dd576c80bb1889408aabd0298f308cd8e9782c1ba5672376ddcf03", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b109e89049ec407b959ceea514fec5496096122cc26c2828e9507b4d3fb835f0", "generation_seconds_batch": 17.235289198928513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.125868320465088, "genuineness": 4.293961524963379, "identity": 0.70630943775177, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cf317b98f3dd576c80bb1889408aabd0298f308cd8e9782c1ba5672376ddcf03.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "f5d8d437f64cb38096180a1da7b2f1f8585114d985d5635e8b1ae4ef75ba5bf7", "generation_seconds_batch": 14.743675475008786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.3641486167907715, "genuineness": 4.331617832183838, "identity": 0.5812184810638428, "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_VOLT_reg_low/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "cf81cfb25ee1663d1b17719b00da1e778d9aef5e4a0cc20d63c6e6b291eaa565", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8070a911a0e901f90aebf69b2ae62b84609bdd74b67afb06e223248a5b6c12e0", "generation_seconds_batch": 21.97868969803676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.038498878479004, "genuineness": 4.356721878051758, "identity": 0.6434017419815063, "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_VOLT_reg_low/cf81cfb25ee1663d1b17719b00da1e778d9aef5e4a0cc20d63c6e6b291eaa565.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "cf9c0ad1a8b263736152a548e27902e65de7cf02df33c80eae2ec99779394f32", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac883498ccee785dff20c5bf9507dfe288a31fabdede8918b150163bb7cbce37", "generation_seconds_batch": 22.66318360203877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.994986891746521, "target_z": 1.9453125391421597, "ce": 6.181467056274414, "genuineness": 4.425758361816406, "identity": 0.6786013841629028, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cf9c0ad1a8b263736152a548e27902e65de7cf02df33c80eae2ec99779394f32.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "cfccd76bad5dae5ca52572730a3f1ddc8197af1e2aff555489a729f48743f352", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64fa8259c56ad38b1b0d4d2d68ae83cae5c1ac30e2483e725ef6d9a7888b581c", "generation_seconds_batch": 21.414761896943673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.507116317749023, "genuineness": 4.507347106933594, "identity": 0.44555556774139404, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/cfccd76bad5dae5ca52572730a3f1ddc8197af1e2aff555489a729f48743f352.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "cfe2f646ccbd7c85990bb077bf1f7ff18fa59659aca0a4197bcfd5c2821d5c74", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d5edc7275e90f0985faa309d1fd152436a4211892fcd0ac2e36e327eef7d1d48", "generation_seconds_batch": 21.085848371032625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7303789854049683, "target_z": 2.250000017158207, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.5100710391998291, "hypothesis": "I found the little box. It was beside the old window. We can look inside today.", "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_VOLT_reg_low/cfe2f646ccbd7c85990bb077bf1f7ff18fa59659aca0a4197bcfd5c2821d5c74.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "d041644b4263c7ea6c2a5780e26aefae7973594be13f8cfe66f747ecf621e20e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c708a2746fb0108531f1c30735c74fa2d5c549d5ed3612168fd8afab40281a9f", "generation_seconds_batch": 19.624494682997465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3342278003692627, "target_z": 1.5546875638070823, "ce": 5.95112943649292, "genuineness": 4.3504462242126465, "identity": 0.7067347764968872, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d041644b4263c7ea6c2a5780e26aefae7973594be13f8cfe66f747ecf621e20e.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "d0e4240caacbf21d4ef78c96156553bd1a71a71f2517d3bf5e48da580b5ada51", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "208e7b343233e6493fe4dd4e1ddb55a795fc98f46e4247beeb87b9575aea2881", "generation_seconds_batch": 20.325472471071407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.008556604385376, "target_z": 1.9296874522787368, "ce": 6.086154937744141, "genuineness": 4.293961524963379, "identity": 0.7234363555908203, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d0e4240caacbf21d4ef78c96156553bd1a71a71f2517d3bf5e48da580b5ada51.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "d10324af9c986ab0a62faf41433398c6dd5e1144863d8cbae35deb6aae7af7f3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "330da5fb4a4fc9fe16258624e2af662aee437ebfa3876f1391f52e31dbd8d82a", "generation_seconds_batch": 16.91272165288683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.0305562019348145, "genuineness": 4.406930446624756, "identity": 0.6987481713294983, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d10324af9c986ab0a62faf41433398c6dd5e1144863d8cbae35deb6aae7af7f3.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "d103fc7842abe8325aab7e3713ee33dbd232a49521c7e54ef518679291267c15", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68c0b8ad5d70fc201f36461ffd0a4af90eb05a3e9ca78d0dc2b28d0a3bf36731", "generation_seconds_batch": 22.56931635295041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 5.689021587371826, "genuineness": 3.7291181087493896, "identity": 0.6877036690711975, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d103fc7842abe8325aab7e3713ee33dbd232a49521c7e54ef518679291267c15.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "d11110c09ef7197f79fa72b40be36c013a9b570c9a0f3a9f53f66150fd25fa6c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07c29848ab81d35b29be91cb48c25fa11cac932d7ea56704603113d10f199cf2", "generation_seconds_batch": 15.597013700986281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3206582069396973, "target_z": 1.5703125134048492, "ce": 6.062326908111572, "genuineness": 4.243753433227539, "identity": 0.7571694850921631, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d11110c09ef7197f79fa72b40be36c013a9b570c9a0f3a9f53f66150fd25fa6c.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "0d0736771afde5d9443193b50d005385f8b2cf6962a57e4bebd8e0405b6dd094", "generation_seconds_batch": 16.014489284018055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.284721851348877, "genuineness": 3.8922953605651855, "identity": 0.22401604056358337, "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_VOLT_reg_low/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "d12bfbad1e0caac4ecc764110f34d8b742f544d65fe53a8fa75e127348f46b27", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "202033afa80fd47b9ce5133c19e0c7bb3a98686daa5aa9e988534a7ed4fc412a", "generation_seconds_batch": 17.247034652042203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.1020402908325195, "genuineness": 4.3881025314331055, "identity": 0.6979154944419861, "hypothesis": "J'ai trouvé la 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_VOLT_reg_low/d12bfbad1e0caac4ecc764110f34d8b742f544d65fe53a8fa75e127348f46b27.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "d16e4deb15b2276b319e17705b7e4137f7bdd0c07aafb19fab682737d98f03c9", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e1b60d7bc40c49c8c9f55c34c17057ea2d9e8b5a16c0ed7c22ec6b35032b9cd", "generation_seconds_batch": 21.863088082056493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.340320587158203, "genuineness": 4.080576419830322, "identity": 0.28535664081573486, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d16e4deb15b2276b319e17705b7e4137f7bdd0c07aafb19fab682737d98f03c9.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": "d199fa9cf88e674546620dee75a036ad19adde547e4a340d1c7badf3607578d7", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d373de6e66b34a218e82170a061fa10f4e77f619d157f75b5c5cdf13995d9a9a", "generation_seconds_batch": 15.206300494959578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.268836498260498, "genuineness": 4.381826400756836, "identity": 0.7658786177635193, "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_VOLT_reg_low/d199fa9cf88e674546620dee75a036ad19adde547e4a340d1c7badf3607578d7.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "d1ab90552216286da1f520f5a4c1d993a520064a74494414cdfee6bc01ca4424", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e2c2454494975b3260d0ff3a330cd9f4ae9b4ce54059c175b4387988be06061", "generation_seconds_batch": 25.15333510900382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.372091293334961, "genuineness": 3.9299514293670654, "identity": 0.5574719905853271, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d1ab90552216286da1f520f5a4c1d993a520064a74494414cdfee6bc01ca4424.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "d1f4bea461385eb7fa9d8f238dfd0077e89b327874a8e77324c52f22e2602b18", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f24ddd877db39f209cb6f16b9d36b4af514ad33f220b94dc069e44cbc424cedb", "generation_seconds_batch": 24.14281178091187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.141753673553467, "genuineness": 4.557555198669434, "identity": 0.7716099619865417, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d1f4bea461385eb7fa9d8f238dfd0077e89b327874a8e77324c52f22e2602b18.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "d204a8d9dc1f45a3f2d1e35dc5eafc252e912ffec9517cae5e9376c812188e68", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38e5b397a3692c44d016aad5db12fcf761ea799233b539002f243c90c80ab2aa", "generation_seconds_batch": 14.900843855924904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.490278720855713, "target_z": 1.3749999571044826, "ce": 6.459460258483887, "genuineness": 4.6203155517578125, "identity": 0.5455324053764343, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d204a8d9dc1f45a3f2d1e35dc5eafc252e912ffec9517cae5e9376c812188e68.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "d210d48bf546ea862641c0baf4b298968bfb3c68ee901fe8ea70f7d5dee07c87", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0d4cafd51108dc43930c290e1b36038cf24d43d4bd7b1915e870e5adaea96c4e", "generation_seconds_batch": 21.223842097097076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.173524379730225, "genuineness": 4.293961524963379, "identity": 0.7217049598693848, "hypothesis": "Ik 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_VOLT_reg_low/d210d48bf546ea862641c0baf4b298968bfb3c68ee901fe8ea70f7d5dee07c87.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "975d04ea07d1e7a9d0e84a2a1f628d838f4e600ee67232883e655e669d207cbb", "generation_seconds_batch": 15.186123343999498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5309876203536987, "target_z": 1.3281249710455256, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.7355157732963562, "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_VOLT_reg_low/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "d2772fe464429c062e44831c6bf13e6f5044fb3accf7bec17453e823ec10b0ea", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e8a13a820cc10ff4a53431f85245d7e6df2e14fda6435cdc6cd0fcc94a0d97b", "generation_seconds_batch": 20.919233610038646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.5459447503089905, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d2772fe464429c062e44831c6bf13e6f5044fb3accf7bec17453e823ec10b0ea.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "d34a9bea7d2c273c208d2e300dc542661655ed60ecbf9a72764cdc566a514f12", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9cabfe98b13230c2f2b93a82223ec9bcf5921ccc35c19e0d20d30f7becdec63", "generation_seconds_batch": 23.21826102794148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.348263263702393, "genuineness": 4.319066047668457, "identity": 0.32391586899757385, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d34a9bea7d2c273c208d2e300dc542661655ed60ecbf9a72764cdc566a514f12.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "d3f15e20ce6a80659eef8ecc565ec69cf04baaf50cc5da7ea33f53a1a5fe505c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "de19e42d3b6fbb6160eca2a5d868235d4f79f395bca98e5acb3a1cfd34a35424", "generation_seconds_batch": 15.188094401033595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.213237762451172, "genuineness": 4.532451152801514, "identity": 0.5295210480690002, "hypothesis": "La porte 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_VOLT_reg_low/d3f15e20ce6a80659eef8ecc565ec69cf04baaf50cc5da7ea33f53a1a5fe505c.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "15062843a7a1b3b502b53bfa371adc008555fbbb982943efd439cd825e0b88ef", "generation_seconds_batch": 16.08366992790252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 5.990842819213867, "genuineness": 4.444586753845215, "identity": 0.7883332967758179, "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_VOLT_reg_low/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "d44e1f6ef7980049e3cf315219f6a7194d5c3676bebe2f6728568b2ed0c2611f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e8969b72d04cedba7d6b7eb67f89d0e1503921bfd8ec57d43711a16265bb349", "generation_seconds_batch": 23.21691831899807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.054384231567383, "genuineness": 4.080576419830322, "identity": 0.3499094843864441, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d44e1f6ef7980049e3cf315219f6a7194d5c3676bebe2f6728568b2ed0c2611f.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "d48dc0aae5167594e22d5f4e68e9238e62e42af256b65bf60142e191a9f8bd1e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a86b181d1df709a1d6030e55eebff1c9733ac95779d28bce4c30d16c222b5b46", "generation_seconds_batch": 16.41944709396921, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5404040813446045, "target_z": 2.468749997855224, "ce": 6.070269584655762, "genuineness": 4.055472373962402, "identity": 0.5847889184951782, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in keren.", "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_VOLT_reg_low/d48dc0aae5167594e22d5f4e68e9238e62e42af256b65bf60142e191a9f8bd1e.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "d4aebece9b00ca99b0262c6dfae876f11d7a1c8b159490d9345607f38c71e3b8", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d1a2be466103c01839fb6401f8ddca6b3d18cd5cc725ea3fbf53ce207e5d449", "generation_seconds_batch": 14.586830905987881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2935189008712769, "target_z": 1.601562549866039, "ce": 6.054384231567383, "genuineness": 4.413206577301025, "identity": 0.5998902320861816, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d4aebece9b00ca99b0262c6dfae876f11d7a1c8b159490d9345607f38c71e3b8.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "d4c11bf79beadc884dc5c415374bb1dcb4c7fa8259ed5d5757e0c49e92c20710", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a76207eff522877c3250ebd4fdb95474bce1c0be09e3c4c1549db27f4bb4aa2c", "generation_seconds_batch": 16.235591333010234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.332377910614014, "genuineness": 4.432034492492676, "identity": 0.38374242186546326, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d4c11bf79beadc884dc5c415374bb1dcb4c7fa8259ed5d5757e0c49e92c20710.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "d4ed7a439d86f078eb8417835db9c15eac15d936913cb22b7d6440727f8c4ffe", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ff9e9913784709a9f59bf944c75f29a3e7b8e5f9a6d8ef4654eba0457729950", "generation_seconds_batch": 16.78196852398105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.229123115539551, "genuineness": 4.005263805389404, "identity": 0.6500887870788574, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d4ed7a439d86f078eb8417835db9c15eac15d936913cb22b7d6440727f8c4ffe.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "d60a9ff93b4bc9a966499b84b922cb7e7764093277fcec839cfec5c7b627af5a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64cdac5b49a7312cd7ac9ee0504ee84f719523d0485b588be5a3a06688ae8840", "generation_seconds_batch": 16.94474713294767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.41180419921875, "genuineness": 4.18726921081543, "identity": 0.38547253608703613, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d60a9ff93b4bc9a966499b84b922cb7e7764093277fcec839cfec5c7b627af5a.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "d632da3c4207639a819a901ace09e7cd55be958a9ccb19a4efc7783df9c9c8de", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "712d4867adb92ade2195791ae0196b7c5b66acb7c5c91d37c6bc302a752b8902", "generation_seconds_batch": 16.257212248048745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.3641486167907715, "genuineness": 4.432034492492676, "identity": 0.617517352104187, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d632da3c4207639a819a901ace09e7cd55be958a9ccb19a4efc7783df9c9c8de.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "d65232e54ba902fe6a5cad8a0dd5eb1c65a4a3d5baa4374d0a2964a001ceaaa0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3b0c129aa1098fe28b58181171b122bf7779f4f5d13548e31dc5b3abb3f4d56", "generation_seconds_batch": 21.45952530601062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0356957912445068, "target_z": 1.8984375530832027, "ce": 6.23706579208374, "genuineness": 3.9550557136535645, "identity": 0.47029751539230347, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d65232e54ba902fe6a5cad8a0dd5eb1c65a4a3d5baa4374d0a2964a001ceaaa0.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "d6fc1f2f132365e19732c152e3fe52cf353752134e76a3c9f2ff30920f5a4e2c", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be407b5ceb9c00cc8e928216aca5c542bcfd9e7f3b114fbb00f47111c93d1a4e", "generation_seconds_batch": 15.856437179958448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.41180419921875, "genuineness": 4.444586753845215, "identity": 0.6470975875854492, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d6fc1f2f132365e19732c152e3fe52cf353752134e76a3c9f2ff30920f5a4e2c.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "d7512a967cdac7cd7dcb6e1737b1c9fe71a194b04c20924e36f40978c8286215", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8907acd9485d1905840d7554707b594a0250be74e6603cc9e12dcf0a1fec7aba", "generation_seconds_batch": 20.78523567505181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.109982967376709, "genuineness": 3.9676074981689453, "identity": 0.7808539271354675, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d7512a967cdac7cd7dcb6e1737b1c9fe71a194b04c20924e36f40978c8286215.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": "d751aa98fda13e503fac763c598a3e48f410bf2556d324e99048de2a91897b70", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3f2dc6fa9d33760dfe3b2eea5a32812481da4d433c76834b2f67d7ffc732f29", "generation_seconds_batch": 21.67341560299974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.38797664642334, "genuineness": 4.2374773025512695, "identity": 0.7030551433563232, "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_VOLT_reg_low/d751aa98fda13e503fac763c598a3e48f410bf2556d324e99048de2a91897b70.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "d76b6ceef62cfaee1e9635a966211deef9e1a58c9b7842fc6f8fa8cc2d09f695", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02459599e6c38e5e5ec4bf709958b229e5dc0ab8ec4c349bcc2646e89cd3f0cc", "generation_seconds_batch": 22.026565699023195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.0305562019348145, "genuineness": 4.218649387359619, "identity": 0.4135053753852844, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d76b6ceef62cfaee1e9635a966211deef9e1a58c9b7842fc6f8fa8cc2d09f695.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "d7f99ff3594232785acdb095ed1f9c0b05bee86b20e654875c667bd5753804c6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27eaf018ff408c5cf30eeb55828570063d2b0adf5dab941f05f7f0f6598b5fc8", "generation_seconds_batch": 21.18937423499301, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.008556604385376, "target_z": 1.9296874522787368, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.6341663002967834, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d7f99ff3594232785acdb095ed1f9c0b05bee86b20e654875c667bd5753804c6.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "d852643668467d65d22708197c51e4fecf0bc367d3920fe3939a4deaa706a5b7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c83626cf497e6c3924dcfe5300fae8a2ab2f2cb48ee068854b8e567dd1016295", "generation_seconds_batch": 21.341055557015352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2121011018753052, "target_z": 1.6953125219839527, "ce": 6.149696350097656, "genuineness": 4.507347106933594, "identity": 0.583768367767334, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d852643668467d65d22708197c51e4fecf0bc367d3920fe3939a4deaa706a5b7.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "d852c6be19bae41b9790d141ddc0e97cee0003f7c253f910acdf9cd0e9df31b8", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e274c97ce3abd75085964680d34b4eb1756cbf179d892eb3443b49bf01f7bb6", "generation_seconds_batch": 16.557278972002678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.038498878479004, "genuineness": 4.093128204345703, "identity": 0.7611845135688782, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d852c6be19bae41b9790d141ddc0e97cee0003f7c253f910acdf9cd0e9df31b8.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "d87b3e9cd8cad4050019d064fe837cd82c6d982d88242351abb2e82f0a1af585", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d3cc18b6e31453ceb79b161d2f37f77b5d01fe2bff461b2bd6ce3a65845bb51", "generation_seconds_batch": 21.51017820101697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.436000108718872, "target_z": 1.437500030026862, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.6726469993591309, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d87b3e9cd8cad4050019d064fe837cd82c6d982d88242351abb2e82f0a1af585.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "d8cab485c7cf85c9f0e95c0b8723b1d3ce9416856c2decfe841822e657c30a16", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "faaf4fdc19f4f5a298898c6c7279980916008eda92fd2db604ca156f05b83ffa", "generation_seconds_batch": 20.851867430028506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 5.839932441711426, "genuineness": 4.268857479095459, "identity": 0.7554470300674438, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d8cab485c7cf85c9f0e95c0b8723b1d3ce9416856c2decfe841822e657c30a16.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "d9149b3babfbd534f2ca71c9b236b21653d567d1736d777bf51ed547b40af767", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c65acad1495ef64f3144eaeb2d071fec82114317e14ddf0287282075b8043e75", "generation_seconds_batch": 20.919233610038646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.062326908111572, "genuineness": 4.557555198669434, "identity": 0.6652522683143616, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d9149b3babfbd534f2ca71c9b236b21653d567d1736d777bf51ed547b40af767.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": "d9703bcb71f09fa3ae68f156919fafd90507f99ddbdd8c71c60e1eeb60e29e68", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "232a535f5d21afc5fde58aac8b4c1a5401f8cc91f5966852f446e32aee89491e", "generation_seconds_batch": 22.691784015041776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.252810001373291, "target_z": 1.6484375359249959, "ce": 6.197352409362793, "genuineness": 4.118232727050781, "identity": 0.6213184595108032, "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_VOLT_reg_low/d9703bcb71f09fa3ae68f156919fafd90507f99ddbdd8c71c60e1eeb60e29e68.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "d986a60f0dc3328c1c15478b1fd6434783f0a2140f4965199a13073d76665bd6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac26e46c94d991268027d95d966d14c583997bb864edc7eecfec501f103c4b56", "generation_seconds_batch": 23.224281108938158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.008556604385376, "target_z": 1.9296874522787368, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.8744745254516602, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d986a60f0dc3328c1c15478b1fd6434783f0a2140f4965199a13073d76665bd6.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "d9af091129aa826e75f06bc361508e2edf89a7e423c05df33feedb337eea3dcc", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58e6e2ead1f1a0fa725d648f401f6881d9996df1173406b2eeac81d6a18b922c", "generation_seconds_batch": 16.79067621799186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.815315842628479, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/d9af091129aa826e75f06bc361508e2edf89a7e423c05df33feedb337eea3dcc.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "d9db7d291cee2e3e43ffc361682b5f261bcc8610863cec9fae3197b242c5a42a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "5651afc7bb81c46072615a93a35f6b5172aab577c4e85eb35da77b9901781042", "generation_seconds_batch": 21.51033430895768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.022613525390625, "genuineness": 4.268857479095459, "identity": 0.6765064597129822, "hypothesis": "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_VOLT_reg_low/d9db7d291cee2e3e43ffc361682b5f261bcc8610863cec9fae3197b242c5a42a.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "da3d247386653c60161a32abe7b19271dc16cfe35e53a6e0c16e98404bb810ec", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43dfefee65584bb071b6cf94b7dff8812dc6c10609b9c6776e7356cc4211f74f", "generation_seconds_batch": 18.652161868056282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.197352409362793, "genuineness": 4.68307638168335, "identity": 0.6420150399208069, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/da3d247386653c60161a32abe7b19271dc16cfe35e53a6e0c16e98404bb810ec.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "dabd597ca51a149bcd4e7714806b8e999d59a9bc1cf516d9b53741da9f804e7a", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3596990f647cc57c10fb382b31c4851c0f797d4921cfba9d312250001eb98a56", "generation_seconds_batch": 21.730016452958807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 5.95112943649292, "genuineness": 3.6663577556610107, "identity": 0.5794164538383484, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás linda.", "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_VOLT_reg_low/dabd597ca51a149bcd4e7714806b8e999d59a9bc1cf516d9b53741da9f804e7a.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "db0756a24e5790aa5e125572f841a510229cfe8883076cc1fe7226f1bfcf6911", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7781eea25a32e406622ea5c51adc60973fa87dd0400de9c191ed347a7674479c", "generation_seconds_batch": 20.948732709977776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.070269584655762, "genuineness": 4.1998209953308105, "identity": 0.6242302656173706, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/db0756a24e5790aa5e125572f841a510229cfe8883076cc1fe7226f1bfcf6911.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "db1197205be90382680929dc6f2d2799bf1d0b760e48df98e19f38152c7a03c4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8233907c8583d9a7a45ada398c800d6150edc8dc74b10be66c5c978cb78e47ef", "generation_seconds_batch": 21.65375940897502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0356957912445068, "target_z": 1.8984375530832027, "ce": 5.959072113037109, "genuineness": 3.9550557136535645, "identity": 0.49250927567481995, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/db1197205be90382680929dc6f2d2799bf1d0b760e48df98e19f38152c7a03c4.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "e581a61194c2b933320a1af756d0b4da1744b30fc7d64d2f30b8ec78a93b5a1d", "generation_seconds_batch": 13.139942476991564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.693823218345642, "target_z": 1.1406250268096985, "ce": 6.403861999511719, "genuineness": 4.319066047668457, "identity": 0.4395129978656769, "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_VOLT_reg_low/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "dc03413bd9b3a8e3fb9e940e8fd48e8c7980e077988a087d7bd611f864b4b5fd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29037f3b0b3796d9566c4fdc0ade11443608b49ea9cfcb533b118fbcd847d4f9", "generation_seconds_batch": 17.065121104940772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 5.847874641418457, "genuineness": 3.9801597595214844, "identity": 0.45036834478378296, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/dc03413bd9b3a8e3fb9e940e8fd48e8c7980e077988a087d7bd611f864b4b5fd.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "dc2650c6c4d70a216c26d5bd14656e8517cae4a5b54561117d6d71d15f1f51a5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0993e6d63cc817cdca7e285c23c76c4f39148d846d380fc4a029c2e7ed52294f", "generation_seconds_batch": 22.733477527974173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8253664970397949, "target_z": 2.1406249581768706, "ce": 6.046441555023193, "genuineness": 4.036643981933594, "identity": 0.6967625617980957, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/dc2650c6c4d70a216c26d5bd14656e8517cae4a5b54561117d6d71d15f1f51a5.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "dc6104df8f194c3186aeb15a20ee1c15d68aa9c1407d8f85e1be3703ce1fbdd5", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09bbbca022ad2653a74363a44c03948f5b23a21e084d957cffa21e09775b60cf", "generation_seconds_batch": 21.245176989003085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.221180438995361, "genuineness": 3.9173994064331055, "identity": 0.3285822868347168, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/dc6104df8f194c3186aeb15a20ee1c15d68aa9c1407d8f85e1be3703ce1fbdd5.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "dd1278a6e9c630a4c8509631988225f587b65207987fd815a8d0f36fd498d897", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "362f5c58b96f58de38ed943a464612d3578f782b21b9cd4f1dbbbff8202b387d", "generation_seconds_batch": 20.325472471071407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 6.141753673553467, "genuineness": 4.607763767242432, "identity": 0.5151534080505371, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/dd1278a6e9c630a4c8509631988225f587b65207987fd815a8d0f36fd498d897.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "dd2e6e0b68a76d5f009ca426d27eb2052857c97e67690e60171b6096cad26dff", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "82560c92038b2ae54c97c56995265ad21538df07115b5381ae01d78d95f17453", "generation_seconds_batch": 22.5631639380008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.38003396987915, "genuineness": 4.369274139404297, "identity": 0.39998769760131836, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/dd2e6e0b68a76d5f009ca426d27eb2052857c97e67690e60171b6096cad26dff.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "dd37e234c0ce72b42b9063cc147f6a3534b1977eac83870933d3405f0a34810b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdad2bcc9d3af0e88a9ee3467824968fd931afdf05cd996b7f0f248555c8dd7d", "generation_seconds_batch": 20.96434709301684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.316492557525635, "genuineness": 4.268857479095459, "identity": 0.5671367645263672, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/dd37e234c0ce72b42b9063cc147f6a3534b1977eac83870933d3405f0a34810b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "ddc3da68de59d9f9f781bb9850060a77e307612bb3c148c211d60a9e86af588f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fb00277c6f46ff4a5efc559e0055e8d39da49e3000786a8aa387d6b0f5873cc", "generation_seconds_batch": 15.496189635945484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.006728172302246, "genuineness": 4.168440818786621, "identity": 0.7685360908508301, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ddc3da68de59d9f9f781bb9850060a77e307612bb3c148c211d60a9e86af588f.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": "de1483167c8fe2ba6c4adeb7f904bb635d1b5e80346ee62f36c47eb034ddd8bd", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Your message arrived after I had left.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e72e0504b484529b5439cd4b6ea833546858e540614ef4a38cc55f6fda187d04", "generation_seconds_batch": 22.300867998041213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.41180419921875, "genuineness": 4.293961524963379, "identity": 0.4446621239185333, "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_VOLT_reg_low/de1483167c8fe2ba6c4adeb7f904bb635d1b5e80346ee62f36c47eb034ddd8bd.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "de3af0494167524201286217a968abcc1217a9391c277ee8e245e395d4b30d4b", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5a1ce301a99a590de21aff22bbfe66f35daa72c1bdc891b6f4af7234e6f9ea2", "generation_seconds_batch": 16.09410828293767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0356957912445068, "target_z": 1.8984375530832027, "ce": 5.808161735534668, "genuineness": 4.055472373962402, "identity": 0.6482541561126709, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuando 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_VOLT_reg_low/de3af0494167524201286217a968abcc1217a9391c277ee8e245e395d4b30d4b.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "de803c284f31cceed051e59d16c5741654b2fe3475149837a2569543d3106627", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd2cfa35e90bfe2b00e439e7b76d370b284f41eef5dabdadc278b761e94dc142", "generation_seconds_batch": 20.262846903991885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.316492557525635, "genuineness": 4.174716949462891, "identity": 0.22439368069171906, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/de803c284f31cceed051e59d16c5741654b2fe3475149837a2569543d3106627.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "de9f7b4bb85cb53fec2cc0cccf8fff178c00627a7efb2c8cbecfb577c91f1e37", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "aef0c8b8ddd28e614d4358ded30b37b4309eb515a3c7d2a95b0d694363fd4daf", "generation_seconds_batch": 15.695639044977725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.157639026641846, "genuineness": 4.042920112609863, "identity": -0.007370941340923309, "hypothesis": "The 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_VOLT_reg_low/de9f7b4bb85cb53fec2cc0cccf8fff178c00627a7efb2c8cbecfb577c91f1e37.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "df07e3509da6b5b50ee5fcf06bad75f9984ea3b4cec9183f686a67dec66fd041", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "654508f9afa9f8b2e6536d350b08b87407290c31ea303e31995a238cba9251ad", "generation_seconds_batch": 15.803445331053808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.165581703186035, "genuineness": 4.105680465698242, "identity": 0.2879156470298767, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/df07e3509da6b5b50ee5fcf06bad75f9984ea3b4cec9183f686a67dec66fd041.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "e03456f19b034b532fad6e9a04cc45bf1037f535248735c8f12cbb612aaf4947", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d75c185b831f6c4b6f61fe934808282a19312551286468622f94ef4bd3f8e921", "generation_seconds_batch": 17.075972283957526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4088608026504517, "target_z": 1.468750066488052, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.13700389862060547, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e03456f19b034b532fad6e9a04cc45bf1037f535248735c8f12cbb612aaf4947.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "e0904e8f3cb218c739ea44f44a54f581d8a1f6d4dcc3a6bdca07602f96a0decb", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27913db28b87c30655b9ff260921df42cf0a34ed04c8ce576ae8e3cf48004462", "generation_seconds_batch": 16.276662845979445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.252951145172119, "genuineness": 4.319066047668457, "identity": 0.5892437696456909, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e0904e8f3cb218c739ea44f44a54f581d8a1f6d4dcc3a6bdca07602f96a0decb.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "e0e12b45782078e5215cf23170ffba1d5b6b36d2bb3574d89cc2962393724755", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d9a5dc509717c6fab0d4b69a2e907bc58d5de26fc38f9500522b148fd727385", "generation_seconds_batch": 21.63971275696531, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 5.982900142669678, "genuineness": 4.2249250411987305, "identity": 0.8150737285614014, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e0e12b45782078e5215cf23170ffba1d5b6b36d2bb3574d89cc2962393724755.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "e10d5da891f97bb08be3949d4eb02b645389cba9dccb7fd4075607adb92a0f06", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "505cccee8c045ce304bc7f10388eff5a6eb61bc902d661ff65a23c4477d185ac", "generation_seconds_batch": 21.95925581397023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.117925643920898, "genuineness": 4.13078498840332, "identity": 0.7148345112800598, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e10d5da891f97bb08be3949d4eb02b645389cba9dccb7fd4075607adb92a0f06.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "e12b1447564e76b40f401cd13fda5d6fbb144339a523f11f98bbf9c7e9175ce4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d53ffd85322dd6b5304748bec25119687f694eaa5634c9c9b79bc4e3d22ed6ab", "generation_seconds_batch": 23.21617014601361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.284721851348877, "genuineness": 4.218649387359619, "identity": 0.6355684399604797, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e12b1447564e76b40f401cd13fda5d6fbb144339a523f11f98bbf9c7e9175ce4.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "e164679f10bf3afe0c821ba70df965e20e761650d4b007809287182812f83160", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5e36c878a12d8e28b348c2bc5e7343c6100a6d19becaeda4c814a276c5a1cbe", "generation_seconds_batch": 22.105166577966884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.197352409362793, "genuineness": 4.570107460021973, "identity": 0.8333566188812256, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e164679f10bf3afe0c821ba70df965e20e761650d4b007809287182812f83160.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "e1779ae5e84d6884ec6818a93c8149ed70ea9ddd09a1fc3b6478f5bd60753556", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efbce791176a5a001511b7b3ac87b3c6d5cfd363dd6c15b50093aae63cca286a", "generation_seconds_batch": 19.142784680007026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 5.982900142669678, "genuineness": 4.105680465698242, "identity": 0.6570042967796326, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e1779ae5e84d6884ec6818a93c8149ed70ea9ddd09a1fc3b6478f5bd60753556.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "e2151d3b7c40838a9f846f1374fb04effc7847eb3edefa883bd99a4739dabe44", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "074eeda91070f6459c8c25e5ed3df61209850df0897ddf63c7d41bce94c8bf4b", "generation_seconds_batch": 22.45217899302952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.125868320465088, "genuineness": 3.9048471450805664, "identity": 0.32377153635025024, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e2151d3b7c40838a9f846f1374fb04effc7847eb3edefa883bd99a4739dabe44.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "e25a5df1f3b31a689c78f206ea7f731c7eb11fbabe2209f0e82abf7d5439e8c1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c0f2305df705b4cc077f92178e35c5aaddce38908d548c9625231faf0bb4fa58", "generation_seconds_batch": 24.195031918003224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.197352409362793, "genuineness": 4.369274139404297, "identity": 0.5887452363967896, "hypothesis": "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_VOLT_reg_low/e25a5df1f3b31a689c78f206ea7f731c7eb11fbabe2209f0e82abf7d5439e8c1.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "e29adda0db958a77f49f4d4634163be5c1f814fd31623cb5238335a87f4d6345", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "109c9fe237456e5c93cd5e0a1ec9d3940c7d6152490e903702b92ca67d0af8e9", "generation_seconds_batch": 23.485427560983226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.086154937744141, "genuineness": 4.381826400756836, "identity": 0.6175328493118286, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e29adda0db958a77f49f4d4634163be5c1f814fd31623cb5238335a87f4d6345.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "e2a69dd64a9cfce650f7134a2ac1ac71637f2ef5488a1524cee09eb77e067119", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4851de9a084bd490077c105733f774abf208b8ec1c88d8958378e56973581841", "generation_seconds_batch": 15.520435606013052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.284721851348877, "genuineness": 4.68307638168335, "identity": 0.6728684902191162, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e2a69dd64a9cfce650f7134a2ac1ac71637f2ef5488a1524cee09eb77e067119.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": "e308ad6941509b563cfae50a3acae6a62fb3cc436d70b4835c64dbf7bfa01a61", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf28954f689e56db310b7f1af9985a7af18f2cc4fc5a3ad3db219ce663c373ab", "generation_seconds_batch": 15.71626257803291, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.213237762451172, "genuineness": 4.218649387359619, "identity": 0.8204377889633179, "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_VOLT_reg_low/e308ad6941509b563cfae50a3acae6a62fb3cc436d70b4835c64dbf7bfa01a61.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "e37eb6a6a49134b7515b4705b071870564d8bbf56f5db460312e1feee3e29ae6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98c2f18b5324a8b5991927ca730d420ed0496fbe0aec47e64c76f56ec83eb4eb", "generation_seconds_batch": 21.171466596075334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.1894097328186035, "genuineness": 4.055472373962402, "identity": 0.33087456226348877, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e37eb6a6a49134b7515b4705b071870564d8bbf56f5db460312e1feee3e29ae6.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "e43eff2343e09e70fef5eaaf02d41ac4cd43d80244c9ffe471eed04df8997e1d", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a53ff9e8e907d6bb759fedfdf888d0189d1ddacf67ff5e953cc683d1d25d38d8", "generation_seconds_batch": 15.823167231981643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.8240814805030823, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e43eff2343e09e70fef5eaaf02d41ac4cd43d80244c9ffe471eed04df8997e1d.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "e444e3d42b6dcbaa852584952b28e17848c3b74204aaf792ec9a272ad8078c04", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fa8e07aa86be1c679ed88ada9aaa8c58b4413c54d7b3b0e86ad3d4e4fef1d1e", "generation_seconds_batch": 21.171466596075334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.260893821716309, "genuineness": 3.936227321624756, "identity": 0.3247694969177246, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e444e3d42b6dcbaa852584952b28e17848c3b74204aaf792ec9a272ad8078c04.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "e44a59405b6970b3060683a768bf8076187b0e7e5fa8ab5a17430b4cd6f328e4", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fb65fab30ae407f9f85040f4e518e8aa2e40a742decf8860a43f5bf351affae", "generation_seconds_batch": 22.313371057040058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 6.038498878479004, "genuineness": 4.381826400756836, "identity": 0.6704967617988586, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e44a59405b6970b3060683a768bf8076187b0e7e5fa8ab5a17430b4cd6f328e4.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "e4c1f155ddd7bd6ef48659c76cf8f1c54bb7d892dd62e1d3186428ce2a91f45c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0e15e0c57da9192f89b40a3269a7dff92471ba27d138886f4312bed17a6303b", "generation_seconds_batch": 22.9904407360591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7982271909713745, "target_z": 2.17187499463806, "ce": 6.165581703186035, "genuineness": 4.149612903594971, "identity": 0.48717671632766724, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e4c1f155ddd7bd6ef48659c76cf8f1c54bb7d892dd62e1d3186428ce2a91f45c.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "e50417fbade253e2ae666b53aa99b15f79d6cc7f1b93d3aad8513c61c0f5909e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b547528fa3b8a0e77366a388977b40c55c399f2779a2c66c12c2526eef833a4f", "generation_seconds_batch": 15.033456400036812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.372091293334961, "genuineness": 4.3504462242126465, "identity": 0.5255023241043091, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e50417fbade253e2ae666b53aa99b15f79d6cc7f1b93d3aad8513c61c0f5909e.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "e558673f0e52c1249cda39c2c0520343a3e638903b7253cd0d0e3fdd64a75830", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "329f97c9c429423d0776110a84288c4f9f7d80626f520cbd79193873bb5e69b2", "generation_seconds_batch": 18.652161868056282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2935189008712769, "target_z": 1.601562549866039, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.6001705527305603, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e558673f0e52c1249cda39c2c0520343a3e638903b7253cd0d0e3fdd64a75830.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "e5a1e26a1517a98f09a8001d84c0be9e17072287655b97988f1c022274299cd3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57c73eb0efdf3f93ece90d1ba608045e52ec6e3acc0655092688a3546ff0ec21", "generation_seconds_batch": 22.55817770701833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 5.649308204650879, "genuineness": 3.9927117824554443, "identity": 0.6710255742073059, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e5a1e26a1517a98f09a8001d84c0be9e17072287655b97988f1c022274299cd3.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "e6742f269d4755f30031371b89b3f3374c36e2b740b91fe5f71b8d04f86f2f74", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a30aab124ef3c1c9b6dd676430390414f5607b0c7175274605da2eaece7f94f", "generation_seconds_batch": 16.00369723595213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2392404079437256, "target_z": 1.6640624855227628, "ce": 6.09409761428833, "genuineness": 4.394378662109375, "identity": 0.7555475234985352, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e6742f269d4755f30031371b89b3f3374c36e2b740b91fe5f71b8d04f86f2f74.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "e678071b4c6511dc067b58815d98b2fc1e23ff1e116d7b463f2a7d67f73ce055", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2e64e33dca3688a598220b72f7ceeb3d2f3004dda1767f752c4a072a688bfbc", "generation_seconds_batch": 15.83743646601215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.266379714012146, "target_z": 1.632812449061573, "ce": 6.165581703186035, "genuineness": 4.356721878051758, "identity": 0.7455874681472778, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e678071b4c6511dc067b58815d98b2fc1e23ff1e116d7b463f2a7d67f73ce055.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "e6a5fe1ca38eb13b567cf7662a2624378b0db44d18f6ef55c300a7b362d6e56f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25e0869c4f9c2a47fe0a85c51be0db717e2847630d62a116b1a508e5ad503f83", "generation_seconds_batch": 23.013066195999272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.133810997009277, "genuineness": 4.507347106933594, "identity": 0.7451248168945312, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e6a5fe1ca38eb13b567cf7662a2624378b0db44d18f6ef55c300a7b362d6e56f.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "e6d37e4d1e832f5ba7d017bede4c4535e2e35fd239aa097a1f1006868a8e7b56", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "496c45875657a6a071aee4507b143a685fd62bae248d56b656a3b40127ba7dec", "generation_seconds_batch": 15.258731039939448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.1020402908325195, "genuineness": 4.406930446624756, "identity": 0.6012496948242188, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e6d37e4d1e832f5ba7d017bede4c4535e2e35fd239aa097a1f1006868a8e7b56.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "e6e907f38dfb1a047cf73ac3fdfc0027d95989463960deffdafbca2fa42094ba", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5aff899a8551164b446d80657ecc53ed6a7852b15267ec12e7f2fa29f48a58a6", "generation_seconds_batch": 22.313371057040058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.070269584655762, "genuineness": 4.394378662109375, "identity": 0.6438807249069214, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e6e907f38dfb1a047cf73ac3fdfc0027d95989463960deffdafbca2fa42094ba.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "3e9922d098e26ba9aba0015d1737e3748808158c50d8e383773515ffe0992cf8", "generation_seconds_batch": 15.863717218046077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.443574905395508, "genuineness": 4.381826400756836, "identity": 0.2347663938999176, "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_VOLT_reg_low/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "e7017044e7525767684ca8087d30ce1c2bc74e7f6083289f4628d0f424c2af24", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae0e64f37ccbadc00778e1b7966842ac755515bcf18f99ba3d4bd684074011a6", "generation_seconds_batch": 15.930861535016447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7575182914733887, "target_z": 2.2187499806970172, "ce": 6.213237762451172, "genuineness": 4.375550270080566, "identity": 0.6113872528076172, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e7017044e7525767684ca8087d30ce1c2bc74e7f6083289f4628d0f424c2af24.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "e734ca36c2cce75a5e782687d04059863fcacb421a2efaf571d446638140ca83", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf002c5c3761f56a73ed2e842b6e7fb55e8982730ba0ae05d083bac72a7733f5", "generation_seconds_batch": 15.883716267999262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5174179077148438, "target_z": 1.3437500579089485, "ce": 6.046441555023193, "genuineness": 4.344170093536377, "identity": 0.7446984052658081, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e734ca36c2cce75a5e782687d04059863fcacb421a2efaf571d446638140ca83.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "e73a473c66da6b5b5839829e9be9d18cc057ba605a52c6618be2817601fb24b1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "bef9fd15876bf783d38131da43351eaf668bdb5fce75f48c3d947aa12d1c2257", "generation_seconds_batch": 16.75457545206882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.920353889465332, "target_z": 2.0312500364611896, "ce": 6.133810997009277, "genuineness": 4.406930446624756, "identity": 0.808134138584137, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que ha dicho.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e73a473c66da6b5b5839829e9be9d18cc057ba605a52c6618be2817601fb24b1.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "e7a29361b5ead4abbf05d970a1aada991c5534685931d4ea341d0cae6257dafb", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48d52fcd763344a9dc54bc637ea507dba4974cff6cf6687a29415c25e147021f", "generation_seconds_batch": 17.065121104940772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 5.903473377227783, "genuineness": 3.9676074981689453, "identity": 0.5533136129379272, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e7a29361b5ead4abbf05d970a1aada991c5534685931d4ea341d0cae6257dafb.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "e7d399610a7ddbc332ae17683423fce0cda57cf049eef3197e6f2f205cc9eec8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ceedc439adeca03e44191479f9d757e7fd57a46cc6a1b0c277b5320e494dcdf", "generation_seconds_batch": 16.026584330946207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.5900462865829468, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e7d399610a7ddbc332ae17683423fce0cda57cf049eef3197e6f2f205cc9eec8.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "e811b2cad4b3604073447d3983560dc362dee70717a9f8e838fa78f9aa9dc215", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a67ef1f92702f7f78122fe542d12bbcaa93420d793c8ea7eb684585b306f88f", "generation_seconds_batch": 17.010945720016025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.933923602104187, "target_z": 2.015624949597767, "ce": 6.006728172302246, "genuineness": 4.532451152801514, "identity": 0.8110817670822144, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e811b2cad4b3604073447d3983560dc362dee70717a9f8e838fa78f9aa9dc215.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "e81326d4476ad888329837009502ccb4b4c3b5e8d87971b88cacd83bce6f2098", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2863ae9e1ffc811eee9f463e1a73c446d73a727c80d317e393906baef406a79", "generation_seconds_batch": 15.860593963996507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5309876203536987, "target_z": 1.3281249710455256, "ce": 6.403861999511719, "genuineness": 4.7081804275512695, "identity": 0.6048687100410461, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e81326d4476ad888329837009502ccb4b4c3b5e8d87971b88cacd83bce6f2098.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": "e87ac4538dcb329989741f5445374d7c3da1bfda0274b98b9e28ae4d01bba5ae", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f5df83fa28e30612132c5008d7fb964c60d741fa3b7630e6ce8450e1e38f9620", "generation_seconds_batch": 17.054749927017838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7710878849029541, "target_z": 2.2031250310992503, "ce": 6.062326908111572, "genuineness": 3.998987913131714, "identity": 0.7360265254974365, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e87ac4538dcb329989741f5445374d7c3da1bfda0274b98b9e28ae4d01bba5ae.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "e8d31f4630764278f466c30a4a6f2250a5cceb94d4089032d94b9b14d9cd50d6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eaeb4de466f07becebc42e09bf154adda2bf20ade35ad1632ed99cb6fb7e848d", "generation_seconds_batch": 15.154782445984893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.006728172302246, "genuineness": 4.118232727050781, "identity": 0.6844223141670227, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e8d31f4630764278f466c30a4a6f2250a5cceb94d4089032d94b9b14d9cd50d6.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "e8d8cdd88d071028bc55f5372401f5071fba94ff1fc2c32a73e319e89c47bd97", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14be7841b1de26290641f0ff90bbc8232e3e2f52e2559d166ee57a67a0da70a0", "generation_seconds_batch": 15.416156007093377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.213237762451172, "genuineness": 4.074300289154053, "identity": 0.6778201460838318, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e8d8cdd88d071028bc55f5372401f5071fba94ff1fc2c32a73e319e89c47bd97.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "e92850a1da084dbb5f618ecd957a2d9f3f1f1ee2a15aeedd07309a917cbf2de1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0c537a1b2b9f72d57a81412449e13ccb13d746e41b6a07be62d03f699e90ac17", "generation_seconds_batch": 16.744034443050623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.341012716293335, "target_z": 1.5468749517425429, "ce": 6.181467056274414, "genuineness": 4.444586753845215, "identity": 0.7477765083312988, "hypothesis": "La porte 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_VOLT_reg_low/e92850a1da084dbb5f618ecd957a2d9f3f1f1ee2a15aeedd07309a917cbf2de1.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "e9510375cf6ded285d737d04eed1ba7d8095daa5c30491d28e00315440b521ed", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8469237ecb0a0af2965703e2301bff53588b9c6ed3a8c5fd473dfdfd197121b", "generation_seconds_batch": 21.434035498998128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.5849686861038208, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/e9510375cf6ded285d737d04eed1ba7d8095daa5c30491d28e00315440b521ed.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "b8c93f8f06b7b15ae63388611dd8dcf1f0154633d1274a648095a6de63d9c0eb", "generation_seconds_batch": 14.743675475008786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.427689552307129, "genuineness": 4.444586753845215, "identity": 0.6026182174682617, "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_VOLT_reg_low/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "ea030d7c4c3c230359d71283488e7c58e6029535a7fb1c954b042a9d9fd53726", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e42a1d2ef2a3ffc3b7fc8c1d7a2536af7ace263b5d9d0dea4c5783734a86fe94", "generation_seconds_batch": 15.74778386799153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6395447254180908, "target_z": 1.2031249624664222, "ce": 6.2767791748046875, "genuineness": 4.218649387359619, "identity": 0.7682446241378784, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ea030d7c4c3c230359d71283488e7c58e6029535a7fb1c954b042a9d9fd53726.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "ea8953c2bbb78369a1795a6545ed18aa48facbfb03ba7e747aa57ed1b1894b66", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6735493f6f9f59cc2778c055b261ba879454cb690a7206aff5e867316660855", "generation_seconds_batch": 21.035757326055318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.197352409362793, "genuineness": 4.011539936065674, "identity": 0.4276367723941803, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ea8953c2bbb78369a1795a6545ed18aa48facbfb03ba7e747aa57ed1b1894b66.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": "ea8ea5ec570b09452d34fc5ebf070c8234454c32e8a391e827d3d9d9617c8135", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f803d7f563b772a7f8e6404e19d66164549da92fc22689754094ad0eab0bfe22", "generation_seconds_batch": 18.068065503961407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.221180438995361, "genuineness": 4.13078498840332, "identity": 0.6833401322364807, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ea8ea5ec570b09452d34fc5ebf070c8234454c32e8a391e827d3d9d9617c8135.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "eaba31d21b00d54aa606cefd1c76f54d9ae295e87e2eb63fcce6df65b03dc8b0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2376c1abfa4722aac99c5410e882e077cbb4fbc5992f8644f0795ec4bb4b0d9", "generation_seconds_batch": 23.87584584497381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1306833028793335, "target_z": 1.7890624941018665, "ce": 6.332377910614014, "genuineness": 4.381826400756836, "identity": 0.7810453176498413, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/eaba31d21b00d54aa606cefd1c76f54d9ae295e87e2eb63fcce6df65b03dc8b0.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "eaeb15ff456e949a44da1a6bc83552bb6c694ad29ab5e73e7aac184f68f6e398", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b9f1fbff2f265ee443afa47f1f29f3ec1242a3606c2ca785327ba403eac2824", "generation_seconds_batch": 22.306950291036628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.3641486167907715, "genuineness": 4.256305694580078, "identity": 0.3962463140487671, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/eaeb15ff456e949a44da1a6bc83552bb6c694ad29ab5e73e7aac184f68f6e398.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "eaf6c86d1f2db389708e14010901edf446546a3f8db5167f28d1454b15abd7b0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d6386b20c3cb16d4f26750dfff2537e05e95c571015156b4a49bc5f11a66b34a", "generation_seconds_batch": 23.42620371596422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8932147026062012, "target_z": 2.062499935656724, "ce": 6.054384231567383, "genuineness": 4.055472373962402, "identity": 0.7517092227935791, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que haya.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/eaf6c86d1f2db389708e14010901edf446546a3f8db5167f28d1454b15abd7b0.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "eb604e23e6991f42572173c094c92db5bbc5725ac1ea3a888c900df17484cf2e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d70c468301ba6d92216c1d514c392e3d24091b11e40c7e09ad5d0157f7434ba", "generation_seconds_batch": 21.781117214006372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.38003396987915, "genuineness": 4.457139015197754, "identity": 0.5575296878814697, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/eb604e23e6991f42572173c094c92db5bbc5725ac1ea3a888c900df17484cf2e.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "eb87d6c4b84de1bd5bab891f313d0f6e782766927d677dbe7923a039545322aa", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b249aef392883e5cd1600491d3b348abef923bf35b2e4a1fb85a19c4d89a1ef1", "generation_seconds_batch": 22.26887457910925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.0305562019348145, "genuineness": 4.218649387359619, "identity": 0.46438270807266235, "hypothesis": "Ik heb een kleosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/eb87d6c4b84de1bd5bab891f313d0f6e782766927d677dbe7923a039545322aa.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "ebb3d38323a60350140990ab1c6a00ef8b8c26861a6fe0507788e759bd7bcc0e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd6d714302bbf929c1f06f707db9ca9ca9ea5c34fbccd5ff1c3bb5fc9a388f86", "generation_seconds_batch": 21.345507033984177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7439485788345337, "target_z": 2.23437506756044, "ce": 6.109982967376709, "genuineness": 4.457139015197754, "identity": 0.4951239228248596, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ebb3d38323a60350140990ab1c6a00ef8b8c26861a6fe0507788e759bd7bcc0e.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "027ec3a5af691cfb259c990278f25e4ec5de86e7eade97dfb210b4d7b1d91077", "generation_seconds_batch": 13.594493446056731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.1894097328186035, "genuineness": 4.262581825256348, "identity": 0.7956185340881348, "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_VOLT_reg_low/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "ebec479532fca212764a23f5be4c74186a9c1ce7cef6881f6424c0f6a3dd90b3", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f064c9510c2e8f56e6314eb0f11e4f79a98756c2475fb669bfc108ec47c3cc7", "generation_seconds_batch": 15.153298906981945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.7860854864120483, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ebec479532fca212764a23f5be4c74186a9c1ce7cef6881f6424c0f6a3dd90b3.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "ec085015c15937b233f7d6604a973ce6e49096e0017743595a19d37bf737ae1c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "690747fa88250e6e50797520554d5f7a7f4a5a32ce7e785794cf625973d227b9", "generation_seconds_batch": 21.322221384965815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.403861999511719, "genuineness": 4.174716949462891, "identity": 0.7451939582824707, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ec085015c15937b233f7d6604a973ce6e49096e0017743595a19d37bf737ae1c.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "ecd8b5a0c5a3c0cdfae9ef19f99c6dae5b840447ca92a4f4ac87fe03df6832f3", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff7cda26a060f21504e5c5e0469607f0863f7d0c065868432baf354160e96268", "generation_seconds_batch": 14.928994455025531, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.7693439722061157, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ecd8b5a0c5a3c0cdfae9ef19f99c6dae5b840447ca92a4f4ac87fe03df6832f3.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "ed105992d7f702615e1ca5326a75b4d132de22b1522e989dcaa3f8041852c80a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "117e9fe5aa335fc078625f3f819df326095af731a5b61c16891262994a3f180b", "generation_seconds_batch": 14.929830803070217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.046441555023193, "genuineness": 4.293961524963379, "identity": 0.5786060690879822, "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_VOLT_reg_low/ed105992d7f702615e1ca5326a75b4d132de22b1522e989dcaa3f8041852c80a.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "ed70e10b00ff8e5663aaf858bbb6694f02e82381b9b918ad953f56ea7d050a29", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6180150de22f36f804f1669df828fb63b935fa701566e2149e9c174361a02193", "generation_seconds_batch": 23.08426889893599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9474931955337524, "target_z": 2.0, "ce": 6.332377910614014, "genuineness": 3.9425034523010254, "identity": 0.3465719223022461, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ed70e10b00ff8e5663aaf858bbb6694f02e82381b9b918ad953f56ea7d050a29.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "ee4217f47850c92078f8478cb92af7bfd2303947e9db1b150e205d99d34dcb95", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18e3c82b13efed31892b458d5b19d253b06f1d78424b10cd7a3e54eab0304183", "generation_seconds_batch": 20.409897082019597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0492655038833618, "target_z": 1.88281246621978, "ce": 6.062326908111572, "genuineness": 4.2249250411987305, "identity": 0.5714901685714722, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ee4217f47850c92078f8478cb92af7bfd2303947e9db1b150e205d99d34dcb95.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": "ee4c41998544204d4fcd21dc33abf5013557b2f29399dbc76acbef104eec3944", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "adf2775e20c1c499fa775fb519881e3d5fd7561c4000506b647d2bc79cbc4c21", "generation_seconds_batch": 19.348783218069002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.252951145172119, "genuineness": 4.218649387359619, "identity": 0.7045185565948486, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ee4c41998544204d4fcd21dc33abf5013557b2f29399dbc76acbef104eec3944.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "ef4c6ccf22a5642103d46eccc587774247b0e7c458bd21d327c6b99f270bfa0f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d0e374b4275cd119b2be91895eb8968f2d1103a48b2d4aa45cb7f03aa7e12ba7", "generation_seconds_batch": 17.42820355994627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.372091293334961, "genuineness": 4.18726921081543, "identity": 0.3493831753730774, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ef4c6ccf22a5642103d46eccc587774247b0e7c458bd21d327c6b99f270bfa0f.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "efb53d78c13eb1933d0a33928d1061a682d6bf3ca664a620a4f139df4b344929", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0651ce6d23d00aaebaf3c4fceb1bb9902d4c2f83c2d9a202dbe55be31090fe4f", "generation_seconds_batch": 15.937258654972538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 5.974957466125488, "genuineness": 4.162164688110352, "identity": 0.7042101621627808, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/efb53d78c13eb1933d0a33928d1061a682d6bf3ca664a620a4f139df4b344929.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "efe8d5c2c5ac418f054b567993646b3bdc4a7feee9115696134f7cc4ea3f1298", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1749d0586aec89c01bbd783a7cc646a82d851d7e09098b47bacfab1ff1abf2d1", "generation_seconds_batch": 22.026565699023195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.1894097328186035, "genuineness": 4.256305694580078, "identity": 0.4875519871711731, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/efe8d5c2c5ac418f054b567993646b3bdc4a7feee9115696134f7cc4ea3f1298.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "f0ac7d611ffabe805f68f3f9799239ceb6a96cd049ac4c380c00d002471d34c1", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4504b189f5a3842e026dd05e83ab33fe46e64edf618fd2c13c6deba859cf8656", "generation_seconds_batch": 15.232622993993573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.402076005935669, "target_z": 1.4765625412869354, "ce": 6.475345611572266, "genuineness": 4.482243061065674, "identity": 0.3747937083244324, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f0ac7d611ffabe805f68f3f9799239ceb6a96cd049ac4c380c00d002471d34c1.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": "f142939c0568e0505e6845a274042c8ef86a9ee0f155375322e313ad22f6c1d3", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c459b8cb4cd01d6c8917d8f6393b7a9051a3b8bfd044d49b3a839f0d362fb1f3", "generation_seconds_batch": 21.67341560299974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.300607204437256, "genuineness": 4.13078498840332, "identity": 0.7820122838020325, "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_VOLT_reg_low/f142939c0568e0505e6845a274042c8ef86a9ee0f155375322e313ad22f6c1d3.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "f146b7a8287e4cc32321b6f5f22b3486e78b25ef459aabf482030e5729ce75f8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d4ba42596bf82e5a6a21eebe40125f6c171a2629b38185180b6915b8b32c76a", "generation_seconds_batch": 21.62704359390773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.165581703186035, "genuineness": 4.381826400756836, "identity": 0.5684239268302917, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f146b7a8287e4cc32321b6f5f22b3486e78b25ef459aabf482030e5729ce75f8.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "f15636774ca06e5bbff5f96098888229b29bafa25f340bfb922cc771c8bd1d44", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2a5fa104374053d98b1b50c2d55d4e12979e27ac084a6b1b7a6dceff0563539", "generation_seconds_batch": 21.411216643988155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.221180438995361, "genuineness": 4.545003414154053, "identity": 0.8083095550537109, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f15636774ca06e5bbff5f96098888229b29bafa25f340bfb922cc771c8bd1d44.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "e0a7d9e670d43c8850051f5db903a0bfea567d0276972dc4d9eff93eefd6bf53", "generation_seconds_batch": 14.545881725032814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3681519031524658, "target_z": 1.5156250525470087, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.6024537086486816, "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_VOLT_reg_low/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "f245f18f7cf3f59f34ceb0af8fe18161215c56e37be986554bb8c9cbc8a97c0f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "5a564940972375cf672a17318a1787827d43bfeae25da031bc2e740d8c4c7a6a", "generation_seconds_batch": 23.148771164007485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6896700859069824, "target_z": 2.296875003217164, "ce": 5.871703147888184, "genuineness": 4.055472373962402, "identity": 0.5551987886428833, "hypothesis": "Die Tür 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_VOLT_reg_low/f245f18f7cf3f59f34ceb0af8fe18161215c56e37be986554bb8c9cbc8a97c0f.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "f2493d8b69c16c8521bb2c2835c3932e5471111853ddceda03929dbe6986328f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e5045f9cdfcd33fb94ce2d5b0dbe00b434d42cee363f5f8243d8ca1f504440b", "generation_seconds_batch": 22.269368395907804, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 5.855817794799805, "genuineness": 4.231201171875, "identity": 0.7693946361541748, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f2493d8b69c16c8521bb2c2835c3932e5471111853ddceda03929dbe6986328f.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "f2d8eb6007bbc91a85cb38632ea7cfca3f1b09e5ad7050fa4eccd62b2839e677", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c79b352dac1e79fb40ced7ba65159f6cf12c3733c1b57315c9e6537d7014be4", "generation_seconds_batch": 21.42492289992515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9814172983169556, "target_z": 1.9609374887399267, "ce": 6.300607204437256, "genuineness": 4.024092197418213, "identity": 0.32804232835769653, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f2d8eb6007bbc91a85cb38632ea7cfca3f1b09e5ad7050fa4eccd62b2839e677.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "f3139cc9889324c99a122f6ff7a6c3b5b58fce0f5d87c04c6aafe276da2cebcd", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "249ab34da915ef43a5dcb0c9965a64eee690fcbeb8d7a6c444d9e4a3fdcb020e", "generation_seconds_batch": 20.627102817990817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.229123115539551, "genuineness": 4.086852550506592, "identity": 0.4905664026737213, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f3139cc9889324c99a122f6ff7a6c3b5b58fce0f5d87c04c6aafe276da2cebcd.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "f33f6a3d8dabf515be59154c587f98f1aaf1c21b7d56bb3ee36700ae297bce4a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "370c9f9be1e4fe9006d65292435374a0080bef5b1de116ed4e213b078b9516ce", "generation_seconds_batch": 16.544953971984796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.197352409362793, "genuineness": 4.519899368286133, "identity": 0.7175279259681702, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f33f6a3d8dabf515be59154c587f98f1aaf1c21b7d56bb3ee36700ae297bce4a.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "f37c82e8f59432f5be51ab6c533c206e3a6c4b023b4d3c93f59b5bdad546509b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f1b501fce7218a8aa456e1de9e6aa18d9c6bc8d0d1565729f0bf54493e90f3d1", "generation_seconds_batch": 17.42820355994627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.268836498260498, "genuineness": 4.545003414154053, "identity": 0.31295230984687805, "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_VOLT_reg_low/f37c82e8f59432f5be51ab6c533c206e3a6c4b023b4d3c93f59b5bdad546509b.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": "f3855d03d05608079afd582126980f50f48b986917e923b3a195cefc800a0f23", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be6188e707e3104059021c8bc359dd3492bb572369984c8d1b7fedbbdda21fb6", "generation_seconds_batch": 16.352822346962057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 5.95112943649292, "genuineness": 4.15588903427124, "identity": 0.7914119958877563, "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_VOLT_reg_low/f3855d03d05608079afd582126980f50f48b986917e923b3a195cefc800a0f23.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "f3e60b45644b1321698ec9a35444e334e4d153e6129b7deb3e02933027ba464f", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f459ce6553c6f2790aa4cb7730c393aaf0f2e1748e40cf8fddebc905e9a23222", "generation_seconds_batch": 19.82065737899393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.141753673553467, "genuineness": 3.973883628845215, "identity": 0.31531140208244324, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f3e60b45644b1321698ec9a35444e334e4d153e6129b7deb3e02933027ba464f.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "f426e9222c882861481f7d02cd81a350a3ee1bf5cdcc2135328ec64b25261f60", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41282df5341cd762aa45a1ceeb247b7bd22aa73deacfaa769d28a7071784fade", "generation_seconds_batch": 15.595161181059666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.173524379730225, "genuineness": 4.469690799713135, "identity": 0.7141538858413696, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f426e9222c882861481f7d02cd81a350a3ee1bf5cdcc2135328ec64b25261f60.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "f42d44b837b20efcfc615b219f3a06216608ef91114e61f169d3dd9a36c4e4b1", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1796d9f7af27f3bc563c59610e97434f970e32f8dc10621674a9ed68294a04a8", "generation_seconds_batch": 22.70895272307098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3070886135101318, "target_z": 1.5859374630026162, "ce": 6.300607204437256, "genuineness": 4.2751336097717285, "identity": 0.7206538915634155, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f42d44b837b20efcfc615b219f3a06216608ef91114e61f169d3dd9a36c4e4b1.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "f46b78f08ec99e16977190b5adae25270175484303c39f2effe01dca056d1ce8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6caa617bde1f02cfea886b4839ff68b1d16d5a3449f905fe9a080c42dfa8a6d4", "generation_seconds_batch": 16.02732338407077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9882020950317383, "target_z": 1.9531250139410432, "ce": 6.046441555023193, "genuineness": 4.268857479095459, "identity": 0.732666015625, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f46b78f08ec99e16977190b5adae25270175484303c39f2effe01dca056d1ce8.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "f5d806e4e96f2f193eb7a6e92edd4bb181f03e454e997c56ae92e38860827c4a", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "732724f2685afbb9f590012ed9f445a347087709a57a0a2ab7ab5fca009f65af", "generation_seconds_batch": 16.420651596970856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.141753673553467, "genuineness": 3.8922953605651855, "identity": 0.3084986209869385, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f5d806e4e96f2f193eb7a6e92edd4bb181f03e454e997c56ae92e38860827c4a.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "f60170433940905ca912d174b8a873c425335507df6aed8ef2375ce3dfe51c85", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "627ce8927129443ad5bd1b1578569470ce4a52c7f5bd5491fdb0a2757ff6e2bc", "generation_seconds_batch": 19.633100384962745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 6.4515180587768555, "genuineness": 4.281409740447998, "identity": 0.5818238258361816, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f60170433940905ca912d174b8a873c425335507df6aed8ef2375ce3dfe51c85.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "f63acbdd9bb1b1f1b49d01f3d51b7cc6d7c88fb95208b3fcb67b58bf53e6e57e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "682b40e823dd48deb8b95228c030d61fd9fb7152815960feedb869b0b1105418", "generation_seconds_batch": 20.561676584067754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3817216157913208, "target_z": 1.499999965683586, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.5916224718093872, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f63acbdd9bb1b1f1b49d01f3d51b7cc6d7c88fb95208b3fcb67b58bf53e6e57e.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "f646802e12649166a0ec0460f5c16202afde002a9f76bd0aa40f23779141f498", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4074091810a2dfc2bffdb4c6ce438da6902b390545d703135044362ca230e12", "generation_seconds_batch": 15.154782445984893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9746325016021729, "target_z": 1.9687499635388102, "ce": 5.895530700683594, "genuineness": 3.835810661315918, "identity": 0.6190656423568726, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f646802e12649166a0ec0460f5c16202afde002a9f76bd0aa40f23779141f498.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "f67e7f243712b0db8fc7318c3b0012ca43dd40d2802daaed6cb69174e6aff522", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf6a164069fb94be7884472bcda3c1ccb4aad0f2d1c5d4dc653611904838e017", "generation_seconds_batch": 22.56931635295041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9067842960357666, "target_z": 2.046874986058957, "ce": 5.855817794799805, "genuineness": 3.6851859092712402, "identity": 0.5341050624847412, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f67e7f243712b0db8fc7318c3b0012ca43dd40d2802daaed6cb69174e6aff522.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "f68ee78f84f98b9a78361049be78325d2e76af46facf9b3c20f94fc9158b7fe6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "820837a861115235e617d4b054e66ac1e3dc81cffcd228abaededbcdbbf385da", "generation_seconds_batch": 17.824970457004383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0289109945297241, "target_z": 1.9062500278820864, "ce": 6.340320587158203, "genuineness": 4.13078498840332, "identity": 0.5324345827102661, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f68ee78f84f98b9a78361049be78325d2e76af46facf9b3c20f94fc9158b7fe6.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "f6e1bb04861d78d93c9c114e4020bd08d5e0e10cf534063aca26a94a2694ecf7", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3e5f602eeb6bae71871553a263d25d25c72a8091542eaef954054bd36febacc", "generation_seconds_batch": 23.224281108938158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.7862967848777771, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f6e1bb04861d78d93c9c114e4020bd08d5e0e10cf534063aca26a94a2694ecf7.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "f6fcff0d8b6d7a1dcf63df12028c24ed46e670d358aac75893900dacfc9461d8", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "91a5c378a45e3cb677725003be81b3ad801bf2af4cbfd3e90d31dc8c58e9e341", "generation_seconds_batch": 20.185415750951506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.086154937744141, "genuineness": 4.331617832183838, "identity": 0.779609203338623, "hypothesis": "De deur 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_VOLT_reg_low/f6fcff0d8b6d7a1dcf63df12028c24ed46e670d358aac75893900dacfc9461d8.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "f70bea423698a494aa1c0e01b7a9cb1f2a3192c613e20467f1932a45fd95867e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fd20ea41bba0efbe1a0775fe38ae03d111c7a1c4cd06966016e1df3087cbe29", "generation_seconds_batch": 20.271699463948607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 5.967014789581299, "genuineness": 4.262581825256348, "identity": 0.7808549404144287, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f70bea423698a494aa1c0e01b7a9cb1f2a3192c613e20467f1932a45fd95867e.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "f713e4926a56ef2dbf9f8077ba6a142fb0a8c69b9b1d170685e16512a9e7887c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d53d3746f034d1eebb760eac2cfcea7b4ba4ded3e09705b9365169cc6b896a90", "generation_seconds_batch": 21.016303991083987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1374680995941162, "target_z": 1.7812500193029828, "ce": 6.09409761428833, "genuineness": 4.093128204345703, "identity": 0.6662545204162598, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f713e4926a56ef2dbf9f8077ba6a142fb0a8c69b9b1d170685e16512a9e7887c.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "f74ef530c9bacb72ca3652f54fa150f5ee64a1658330459ef1a0e7e91b68c116", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4a8e74a208f6ecc199caf98a74bdbb1a8d1f420d26aa1aa1c43c1837e8eb250", "generation_seconds_batch": 21.329932671040297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660753965377808, "target_z": 2.0937499721179136, "ce": 6.348263263702393, "genuineness": 4.074300289154053, "identity": 0.6218922138214111, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f74ef530c9bacb72ca3652f54fa150f5ee64a1658330459ef1a0e7e91b68c116.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "f789d0ecf66bd2861fa534f1b69bf7aac68a294064330e66936e1bfc24b49a20", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "589ddcbe110b1220edfe85d3dbbd55e227b64154649ed097c71bc9784bf9389d", "generation_seconds_batch": 17.246131401974708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.09409761428833, "genuineness": 3.898571252822876, "identity": 0.3552769124507904, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f789d0ecf66bd2861fa534f1b69bf7aac68a294064330e66936e1bfc24b49a20.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1c8404c5f4af8978ec620c0c9183f112845b3c651ce9e1cac870e4f5f3e6e269", "generation_seconds_batch": 15.203751761000603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2460252046585083, "target_z": 1.6562500107238793, "ce": 6.062326908111572, "genuineness": 4.356721878051758, "identity": 0.7887621521949768, "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_VOLT_reg_low/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "f7cbc8ee3a4ef863da9fc910de01be204cf28dcb493cd260e3aea7dd9c2904ce", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a6456b483094345c4087a824e486521d452c01ebf41a04347f3ea2b753f619f", "generation_seconds_batch": 23.018887489917688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 5.95112943649292, "genuineness": 4.306513786315918, "identity": 0.7506753206253052, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f7cbc8ee3a4ef863da9fc910de01be204cf28dcb493cd260e3aea7dd9c2904ce.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "f86bd943a4b501c0c34babad70cbc455a7056cc54a65c5ea92c192bd2fb18963", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "24f1b52639bada965198741d9f0d7593a9aea8f2e938d1c765694c3c0fe29f45", "generation_seconds_batch": 23.687269009999, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1849617958068848, "target_z": 1.7265625584451425, "ce": 6.070269584655762, "genuineness": 4.168440818786621, "identity": 0.7799723744392395, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f86bd943a4b501c0c34babad70cbc455a7056cc54a65c5ea92c192bd2fb18963.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_VOLT_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": "82beb1f914d4fc0718e085a10cdb192066f40dc9d3b648216d2076e617d64bdb", "generation_seconds_batch": 16.99135822802782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 6.308549880981445, "genuineness": 4.519899368286133, "identity": 0.6626126766204834, "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_VOLT_reg_low/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "f8b151e45e06141e4eeffd2a5b2f169cea04f0aa103f9b14f68da4ff8f2fa1e0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2012b110436dce44adc0a1d5da0cb5dfa300f497508ec46503930c8655d53ecd", "generation_seconds_batch": 23.000565672060475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 5.919358730316162, "genuineness": 4.093128204345703, "identity": 0.6558874845504761, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f8b151e45e06141e4eeffd2a5b2f169cea04f0aa103f9b14f68da4ff8f2fa1e0.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "f8f3e07fbd2404b6821d5c05609cc0b66ad1bbcf4a4e6be927711651c1cbc62f", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0800ebd2846aa44e841dcd9e31785aeaa7aa74b363d760caec06c39f4a77766", "generation_seconds_batch": 15.834455555072054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.5078495740890503, "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_VOLT_reg_low/f8f3e07fbd2404b6821d5c05609cc0b66ad1bbcf4a4e6be927711651c1cbc62f.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "f9377eaea13267985436c80fe235cac289cae588ffdab22ebebbfe7c4c82831e", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8f94a7cda0269fcd218496b83a182da1195daee4a7d29d06230160fddd0c314d", "generation_seconds_batch": 19.066799959982745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.676100492477417, "target_z": 2.312499952814931, "ce": 5.808161735534668, "genuineness": 3.8797430992126465, "identity": 0.8271881341934204, "hypothesis": "La porte 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_VOLT_reg_low/f9377eaea13267985436c80fe235cac289cae588ffdab22ebebbfe7c4c82831e.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "f9b4d65a521f6811200f39d5934ce09b23ec18864251b8f21d5e8f52d90e354b", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32d41469fb20f46ebce2e470a1f8855cde0f4da46711fd0a6844ebcf1e775bc0", "generation_seconds_batch": 16.44835966802202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5716965198516846, "target_z": 1.2812499849865688, "ce": 6.229123115539551, "genuineness": 4.256305694580078, "identity": 0.5196782350540161, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/f9b4d65a521f6811200f39d5934ce09b23ec18864251b8f21d5e8f52d90e354b.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "fa0f4beccf80a89f82907a43bb7944cb25c26b836ac8c718eb154cdc552227f6", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57528574ab6bc1c3d6415523256ac55b6620cdd450ef1fa6907ce083181336e2", "generation_seconds_batch": 23.000565672060475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2935189008712769, "target_z": 1.601562549866039, "ce": 6.213237762451172, "genuineness": 4.287685871124268, "identity": 0.5322315692901611, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fa0f4beccf80a89f82907a43bb7944cb25c26b836ac8c718eb154cdc552227f6.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "faa6c4cc9fde193a986e32d9756f0cb9c4fae6321fb39745db23562fbdb5cd36", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c16331c1414f4f135c66fbf4591b5ba30226e6b1ea14cf836a7553152bc9c4b2", "generation_seconds_batch": 22.204893738031387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0560503005981445, "target_z": 1.8749999914208966, "ce": 6.109982967376709, "genuineness": 4.080576419830322, "identity": 0.6353280544281006, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/faa6c4cc9fde193a986e32d9756f0cb9c4fae6321fb39745db23562fbdb5cd36.mp3" }, { "id": "vn_VOLT_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-0", "key": "faf56180cb0c39302d3b08dcee0deace32c8144a9435d2376e0361dc981c0e39", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23e03643f2a3fa375695616210e5b6902a436c7a3c1ae012510bc15c418b9616", "generation_seconds_batch": 21.618945914902724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 5.998785495758057, "genuineness": 4.068024158477783, "identity": 0.7633587718009949, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/faf56180cb0c39302d3b08dcee0deace32c8144a9435d2376e0361dc981c0e39.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "fb6bc72be857bfe2faf6273b3f874bdd8b623fe09bc06d4fd31d7371eef7b2ef", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a9dfc74c22b5f305a061c5dcf72827225d2c60d6a248257a58eaf23b0b7493b", "generation_seconds_batch": 19.63305997592397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0153414011001587, "target_z": 1.9218749774798534, "ce": 6.252951145172119, "genuineness": 4.030367851257324, "identity": 0.2806505858898163, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fb6bc72be857bfe2faf6273b3f874bdd8b623fe09bc06d4fd31d7371eef7b2ef.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "fb6c742e64e637f4f717dd561c22dde2a4fbc2c840a4680750f1c6673c0bb778", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45fd6dfa5e2f49c9bcdb916e8001c75909c771b39ecd5084f1141886dfdb6260", "generation_seconds_batch": 22.947837801999412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 5.967014789581299, "genuineness": 4.149612903594971, "identity": 0.3783443570137024, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fb6c742e64e637f4f717dd561c22dde2a4fbc2c840a4680750f1c6673c0bb778.mp3" }, { "id": "vn_VOLT_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-es-1", "key": "fb7270ddde01ccffa5d1b0f5a428d9cc3bad447352c7c79ebaca28d42949cc49", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47ae1279decdfa18d5cb803a75216f413fad6edfe83c8b5bebfc8ac7b02f5907", "generation_seconds_batch": 15.915383031009696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.1020402908325195, "genuineness": 4.086852550506592, "identity": 0.696577787399292, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fb7270ddde01ccffa5d1b0f5a428d9cc3bad447352c7c79ebaca28d42949cc49.mp3" }, { "id": "vn_VOLT_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-0", "key": "fbf17e3766112896114bfbb9d00fc639b149a2d809df5ef99789e1c16f398362", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c0d5d9ab9b38f081ad73ca7cdb2afa336865c06527ff2b4677f8cfac514453f", "generation_seconds_batch": 19.624494682997465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.734532117843628, "target_z": 1.0937500407507417, "ce": 6.2767791748046875, "genuineness": 4.457139015197754, "identity": 0.6924219131469727, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fbf17e3766112896114bfbb9d00fc639b149a2d809df5ef99789e1c16f398362.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "fc060713b593e9a32a97eb898cff430de32d03a207196a3511290c41a563aa02", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "934a032ae5414b01e293d70c5786b6b417fb96b9f08fde97c058c62c0cadb1f3", "generation_seconds_batch": 21.930645839078352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.395919322967529, "genuineness": 4.20609712600708, "identity": 0.6359807252883911, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fc060713b593e9a32a97eb898cff430de32d03a207196a3511290c41a563aa02.mp3" }, { "id": "vn_VOLT_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-1", "key": "fc1d2b0a4a3370c1847d00056b44023fdafa9814fa13df8484527e9ae300119b", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23fabde8004e5b2ac892d3ecdd747a24acef40e1aa0e20b5c8e3c1e1679afb16", "generation_seconds_batch": 16.697846231982112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 6.340320587158203, "genuineness": 4.293961524963379, "identity": 0.30073973536491394, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fc1d2b0a4a3370c1847d00056b44023fdafa9814fa13df8484527e9ae300119b.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "fcf029badc3864f78f942be70a1f535ee1f49498aedd21ba2e93501dc3c61e3d", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6da0a46c18e6a286230253df4fff47366bc1573167b0edb3f4254aff61439e13", "generation_seconds_batch": 22.76585518801585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.205295085906982, "genuineness": 4.519899368286133, "identity": 0.7342133522033691, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fcf029badc3864f78f942be70a1f535ee1f49498aedd21ba2e93501dc3c61e3d.mp3" }, { "id": "vn_VOLT_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-en-0", "key": "fd1f70d6850b38511495af4c6bd79f5b8984df2ea29cfd4359060c8c0a1c209c", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad2f053a56afe9bf05b31f91bdb0f949370289393f54e6451fc71a8f70187b94", "generation_seconds_batch": 17.41362712800037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8796449899673462, "target_z": 2.0781250225201466, "ce": 6.197352409362793, "genuineness": 4.017816066741943, "identity": 0.30377209186553955, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fd1f70d6850b38511495af4c6bd79f5b8984df2ea29cfd4359060c8c0a1c209c.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": "fd8079cb99627aa33d602840a2dbbe70d52c8e35fc20368c32373df40ec0d235", "bucket": "vn_VOLT_reg_low", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05904940300aa5315a6e41a1a7984e7bcaf76f30ef2fa72b7d7779fbc60c9853", "generation_seconds_batch": 22.390626218984835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9610627889633179, "target_z": 1.984375050402233, "ce": 6.221180438995361, "genuineness": 4.720732688903809, "identity": 0.7638888359069824, "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_VOLT_reg_low/fd8079cb99627aa33d602840a2dbbe70d52c8e35fc20368c32373df40ec0d235.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "fda9e5547dc65b5674a89c4d7c7018ab8796b7e5a27e04dab8c20ca33a470f78", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "347018a03a50302c1501fb1c9d2d114a4286f397f7aa5125f92a21b453ebab05", "generation_seconds_batch": 15.854795605991967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5649117231369019, "target_z": 1.2890624597854523, "ce": 6.292664527893066, "genuineness": 4.482243061065674, "identity": 0.6847830414772034, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fda9e5547dc65b5674a89c4d7c7018ab8796b7e5a27e04dab8c20ca33a470f78.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "fdc7119d8bc14eed4b701b8e7624ae5533920deb374c759f85d48919cb6f1564", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33cb5a89094f0328e3bfc5d33df14229e51ba42f73dc27780f6ed055c33476b6", "generation_seconds_batch": 23.369914517970756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3070886135101318, "target_z": 1.5859374630026162, "ce": 6.062326908111572, "genuineness": 4.268857479095459, "identity": 0.7150839567184448, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fdc7119d8bc14eed4b701b8e7624ae5533920deb374c759f85d48919cb6f1564.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "fdd51ddba955073a288e21f20f7a9eb068228dce956d82f5a0924925c2c74aa0", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb044ea807f3635cc9c694df11f9885c84bfad3a7245e37712f3d026735814a1", "generation_seconds_batch": 21.76626543502789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4996950626373291, "target_z": 2.515625121179837, "ce": 6.260893821716309, "genuineness": 4.099404335021973, "identity": 0.6366142630577087, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fdd51ddba955073a288e21f20f7a9eb068228dce956d82f5a0924925c2c74aa0.mp3" }, { "id": "vn_VOLT_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-1", "key": "fef77d2ac77b7bb60a44aad0ac5702ba468e1d254fc56a1f374d1f166c74ad64", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67d5c1e25a5d67ffc131bdd195fd9487d5d3110fb1b51028e706661c8d283bcc", "generation_seconds_batch": 22.105166577966884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8389360904693604, "target_z": 2.1250000085791036, "ce": 6.268836498260498, "genuineness": 4.482243061065674, "identity": 0.7901041507720947, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/fef77d2ac77b7bb60a44aad0ac5702ba468e1d254fc56a1f374d1f166c74ad64.mp3" }, { "id": "vn_VOLT_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-nl-1", "key": "ff534dee853fc3fdeea5caec5af98d60e84fc1c4e0e24a75aad602655b6ac68e", "bucket": "vn_VOLT_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "310a1e245b5dc8fde684110786db970737d98e43df2e1c3abcadc08c49f9bf64", "generation_seconds_batch": 15.8695157780312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 6.229123115539551, "genuineness": 4.369274139404297, "identity": 0.602830171585083, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ff534dee853fc3fdeea5caec5af98d60e84fc1c4e0e24a75aad602655b6ac68e.mp3" }, { "id": "vn_VOLT_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-0", "key": "ff5f1e12e8fff286561cf496d42d8955a4cc11f0566198319da9bcd1be363103", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e07e9f2c783603b9d5944b5c9b98e0810239be9c512ffc7e6a68e2adacfce42f", "generation_seconds_batch": 20.84800118894782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.23706579208374, "genuineness": 4.231201171875, "identity": 0.5893065929412842, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_VOLT_reg_low/ff5f1e12e8fff286561cf496d42d8955a4cc11f0566198319da9bcd1be363103.mp3" }, { "id": "vn_VOLT_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-de-1", "key": "ff75d015baa6e3fa8d4401fa95ad6e286a27c21ee40edcd4078f5c5c6d5463ed", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1be365702a2a7ab777f2bcecf87d9d4a88468bcd59562a03a39f825b02781fe", "generation_seconds_batch": 23.21617014601361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8525056838989258, "target_z": 2.1093750589813363, "ce": 6.284721851348877, "genuineness": 4.344170093536377, "identity": 0.6907668113708496, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam erinnern.", "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_VOLT_reg_low/ff75d015baa6e3fa8d4401fa95ad6e286a27c21ee40edcd4078f5c5c6d5463ed.mp3" }, { "id": "vn_VOLT_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_VOLT_reg_low-fr-0", "key": "ffb933068518812e83f49f490466e987a94e406a0f1f7d18fc1f0209052ab9ad", "bucket": "vn_VOLT_reg_low", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Emotionally stable and steady delivery, maintaining a consistent state while allowing natural expressive phrasing. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0889eb694fae12a572dc48d3220363ce8e52e885e18745835dcc446ca94a26f4", "generation_seconds_batch": 21.76490723795723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.8224508166313171, "hypothesis": "La porte 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_VOLT_reg_low/ffb933068518812e83f49f490466e987a94e406a0f1f7d18fc1f0209052ab9ad.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "b287df82c12eaa0dfa7b119cbcc612776bf02cad73a2dd5afa4595864731a60d", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.35272741317749023, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/3cbd0af6cd9e514f327c3f07.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "9016cc9f2fd209bb2faebf68316b5d4722ec4fb9e3b854638e6cfefc7443e9ce", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.181467056274414, "genuineness": 4.369274139404297, "identity": 0.5142604112625122, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/d6a319eb714c5c9bfcf84bdc.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "946e1296096420a32c90cb2c3fe3f3228ea94a96eea14cdd287c87a7c0194039", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.36587265133857727, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/62587111aa1362a891cfe3d4.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "30b8997ec94cfac0b7b02bd775d64023928d592bcab93de9a7eaadcea0007ea2", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.149696350097656, "genuineness": 4.381826400756836, "identity": 0.5704646706581116, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/d7ac24fe01b7eb05d59cf69d.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "8bde3965c7edf3067c334b37f813e554ec4d3edb0caa5ec3b0bd49f5f74e5d6b", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3206582069396973, "target_z": 1.5703125134048492, "ce": 6.165581703186035, "genuineness": 4.570107460021973, "identity": 0.7217540144920349, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/a8c71db747629b98ecc6c80e.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "f0bf0e4ed4b44ecfeea62bfa956b86bcee102a539652f5f70baa0cf6ebfc3afc", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.402076005935669, "target_z": 1.4765625412869354, "ce": 6.1894097328186035, "genuineness": 4.6705241203308105, "identity": 0.32765722274780273, "hypothesis": "Die Tür ist noch offen? Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/e814a88917ac5a35a7b45c8a.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "b07f91f25001fb6214bef23f38c06e22227603779e432e6202233efafeb6609e", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.284721851348877, "genuineness": 4.557555198669434, "identity": 0.5142760276794434, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/75cf019316307e3cb733cec6.mp3" }, { "key": "original-vn_VOLT_reg_low-de-0-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "98cc398071d38af487efdca5e633565228e821fef79708e98405c0cd77e23d8c", "task_id": "vn_VOLT_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2256708145141602, "target_z": 1.67968743512053, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.47841355204582214, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/e39daf60e472a6943e2f94e8.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "a0b9da03ffbafa86bf465ccec02b2dafedb855c634ca5de645f95cba18c7fcda", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.266379714012146, "target_z": 1.632812449061573, "ce": 6.300607204437256, "genuineness": 4.168440818786621, "identity": 0.7584918737411499, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/9aeea744d5978aa077a1ec38.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "3f19363719bd4ac7e0eb9c82e8673a56584d8d92f071efdb28374cd2c59852cb", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.38003396987915, "genuineness": 4.3127899169921875, "identity": 0.6208655834197998, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/f5703c3dadbaf596b99ff606.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "3d06769ef5e7f1de76ed408fdab883ea8526a0665f6758a2a84d5b88a1c01bd8", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.38003396987915, "genuineness": 4.18099308013916, "identity": 0.4382786750793457, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/5b1d1deec82bcff1005e819e.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "f0388b6325f302f7470fb218d15a889f4fa61de4209c0bfe55057c58a6b7f198", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.332377910614014, "genuineness": 4.149612903594971, "identity": 0.48284924030303955, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/b522a3d08c7644a6288e5c05.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "8dbaea604782b3f517045c74876f37af8b29a47f7009c865e50bfea4340e3cb0", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.348263263702393, "genuineness": 4.381826400756836, "identity": 0.7058544158935547, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/fb6125661f9a474e4032b048.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "4a7488f8563bf912581f72e1473438f144179a6e8173e4f134cf331e83a84802", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.38003396987915, "genuineness": 4.243753433227539, "identity": 0.5025814175605774, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/f60971582c58c46ecf777768.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "ce16e14546a60c161736b4b9e48d33d67ec1860521d27e6c12208191764bff60", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.41180419921875, "genuineness": 4.293961524963379, "identity": 0.522524356842041, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/4a7639e3afd35ae115d4fada.mp3" }, { "key": "original-vn_VOLT_reg_low-de-1-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "a1c20ae27309f80cdc9b200de136e0e3cf467301a93c23a7c42fe6704c43bc1b", "task_id": "vn_VOLT_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.252951145172119, "genuineness": 4.118232727050781, "identity": 0.5626938939094543, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/de/80b68a1f47b6dbf5a05b18ec.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c06108592bca816664d98630c9b98c6563c11635aa9434406a61b338e2d164a4", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3477975130081177, "target_z": 1.5390624769436594, "ce": 6.2767791748046875, "genuineness": 4.319066047668457, "identity": 0.3078502416610718, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/68fbbc77299a1ac8dfc4db5d.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "aed5e7705853686b88690c7156298ef1eff29ace4c33faeda41e3d51a14ba73e", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 0.7846575975418091, "target_z": 2.187499944235827, "ce": 6.125868320465088, "genuineness": 3.8922953605651855, "identity": 0.2964863181114197, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/12d837193df81bf62f5533fd.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "1a49983e114978a952a0ec25e8b19e8ea6fedd49610f5eaf59c01421a26e45f3", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.284721851348877, "genuineness": 4.042920112609863, "identity": 0.3242532014846802, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/b55f98af5d9170591ea9faf3.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "bc2dad8042b7107a417548a44fd1480b522b203177c988db6def18ddfe4bb8de", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2867341041564941, "target_z": 1.6093750246649225, "ce": 6.046441555023193, "genuineness": 3.9927117824554443, "identity": 0.19826701283454895, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/9f58d87a2e4aaf170d6d0289.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "ee6cc749c23d5eb61d146308aad87e6931b6b14a691402b2e951feedf4e0f7d7", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4224305152893066, "target_z": 1.4531249796246293, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.2047388106584549, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/1e5152e055e300dd98fd6f38.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c92240b167580d3cfc4924f6015539e503c64c3a9ed9d04f736e334afe8dbed6", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.38003396987915, "genuineness": 4.124508857727051, "identity": 0.177343487739563, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/189f31638a8daf1017a8893e.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "b738e7906397512e8bda0c61d1280f3b6f05829436dd2a708dd8d7f0e1f70106", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.165581703186035, "genuineness": 4.055472373962402, "identity": 0.009843224659562111, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/a7169ba4ad7baa3eb0567bcb.mp3" }, { "key": "original-vn_VOLT_reg_low-en-0-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "ade260988679ac6170f7a058ae5821f83e87923683705979b951e654e1669674", "task_id": "vn_VOLT_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.213237762451172, "genuineness": 4.2249250411987305, "identity": 0.13419382274150848, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/c2ba9445cbd61eb78b17bef4.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "f13d78252c11738eb17597bc36b807c892040e9dac6894cecb17823edfec7640", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5309876203536987, "target_z": 1.3281249710455256, "ce": 6.41180419921875, "genuineness": 4.482243061065674, "identity": 0.33174020051956177, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/56f99ab49cc2c2a61f67920e.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "5779ebf1c6cadb03a7bf3e240ad8c164b4f7c72418711ce98ae471661ae47e79", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.403861999511719, "genuineness": 4.482243061065674, "identity": 0.3473796844482422, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/d936c6c003f9eb69cbd45ccd.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "f22bac24a7aa8643e4982c02487fae0070b23a38e6b569841f49d63ba3a1fd91", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.19647060334682465, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/447462e1972e15eb94e590a7.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "1c635bc3acf559680466a35b79ba72d2fe59fc1f3cf19ebedae08f03268327dd", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1985315084457397, "target_z": 1.7109374715817196, "ce": 6.38797664642334, "genuineness": 4.268857479095459, "identity": 0.4359072148799896, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/39bbadea00649ae336f56511.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "0f01674bbc5db591317ea24d5319090558518897e32a04293326e74529afa945", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4970635175704956, "target_z": 1.3671874823055992, "ce": 6.4515180587768555, "genuineness": 4.394378662109375, "identity": 0.36434096097946167, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/71c5c541639bfc4de6ee9010.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "a7a1f47f10678e0be5c793a75b1c078bfbfa6ce41248293e15caabbea3053523", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.38797664642334, "genuineness": 4.344170093536377, "identity": 0.3561251759529114, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/381161549bf2f7f64207f199.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "af58b03feb3d5cc989b9d397d653ccf4ef845796c5dab1d4558e4c2df3b7605d", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5445572137832642, "target_z": 1.3125000214477587, "ce": 6.2767791748046875, "genuineness": 4.281409740447998, "identity": 0.32582905888557434, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/47e5a0cc05f76a4b8c4e4e0c.mp3" }, { "key": "original-vn_VOLT_reg_low-en-1-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "e85fed313e9503e1c5ef629370104e952acfeb89d9ef7a7c7e187c6a2bd7bfdd", "task_id": "vn_VOLT_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.41180419921875, "genuineness": 4.300237655639648, "identity": 0.39755964279174805, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/en/272d7af1b9716cad98bd736d.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "a035a9d2af4c3d1a2fbec4ba4a87557e67acd9183eb1a02b3b4a5e71f818d303", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.165581703186035, "genuineness": 4.262581825256348, "identity": 0.6440675258636475, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/2691a6b2d26744f513f5d336.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "4a323f47c2afd52f9db6dfb3c51e9ce31ba326ec221da8ff640a5e2f8610cbe9", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1713922023773193, "target_z": 1.7421875080429094, "ce": 5.935244083404541, "genuineness": 4.469690799713135, "identity": 0.6628453731536865, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/373bed9c4d2e4633b9a5ef1b.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "285752b08c944dc4486a358f561249260ff5ceb77de6a815afb40303a2dfa9a1", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 5.839932441711426, "genuineness": 4.13078498840332, "identity": 0.6334822177886963, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/ebad540891c25259023c3fb5.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "e83202b649c00f46b92654f4b247cb650a3b614d46cbc80b578792352d32f9ba", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 6.0146708488464355, "genuineness": 4.105680465698242, "identity": 0.6291584968566895, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/f8e9d8f193239ad5b45bb372.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "3b5fe8234a4f2b583cecbca56c7162522b0499ed1141a6ebce4de07efb6974ba", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.022613525390625, "genuineness": 4.231201171875, "identity": 0.666542649269104, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/701a48b6f251b993cb20abb0.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "d891a3d051b8bef6a08aa0fe3245ff39a95fc5a82b8d595b84c2441eb21b0d7d", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.205295085906982, "genuineness": 3.8546390533447266, "identity": 0.5221863985061646, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/fb8074c97fef127932834d6a.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c48a8d5058a946a81e87402a370ebacbb08e758efa500df1b66b51c52e8afe99", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.125868320465088, "genuineness": 4.15588903427124, "identity": 0.5908240079879761, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/43246cc183c22abc1679db01.mp3" }, { "key": "original-vn_VOLT_reg_low-es-0-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "e7f123f7d773453bd062381233d9eb9d6f501c52680d8ae2ade7212b945555fb", "task_id": "vn_VOLT_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0017716884613037, "target_z": 1.937500064343276, "ce": 6.1020402908325195, "genuineness": 3.8044304847717285, "identity": 0.6052196025848389, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/2401502887c2ad3e3dfd519c.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "6a1e1139950780b49cdd694f3886748a6b6264f46afd063a8a1205361800c315", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.133810997009277, "genuineness": 4.7458367347717285, "identity": 0.6440640687942505, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/28875da9ec08d0aeb5f4c19d.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "9966205cff8618834f2a51c22c9144a0f6f737bfa50c7e5bd52f4318dea59a96", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.361367106437683, "target_z": 1.5234375273458924, "ce": 6.117925643920898, "genuineness": 4.444586753845215, "identity": 0.7220869064331055, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/4199c41aa608ca2301c99ccf.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c0cd7adb2421653699b0bcb079ccd945e5b57111073aa462a4938679afa77e93", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5309876203536987, "target_z": 1.3281249710455256, "ce": 6.260893821716309, "genuineness": 4.68307638168335, "identity": 0.6266487836837769, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/8d3e95ddcebcf0935229af05.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "7d428757b59b840ff9280d6221376115313959f3f819c6a597bc8796062a24b8", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.117925643920898, "genuineness": 4.582659721374512, "identity": 0.7869511842727661, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/86f2744180c140f1d32e2583.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "258c40d89feaa198f92fd91a6e39c9b804dc1393acc2401e92f611e836e3f073", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.6479090452194214, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/734aeb504983326dc73c9f8c.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "762151c4e4ba3332303eccc63fb0cb36c1fdd8794dad04e436b1cdba39dac3cd", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.06961989402771, "target_z": 1.8593750418231294, "ce": 6.086154937744141, "genuineness": 4.494794845581055, "identity": 0.6705252528190613, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/74312614bd11b6c05ff27ca6.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "2db87a2fcc87610b716faacd50a6b8b68fedfd5980b179673d69269843344caa", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.181467056274414, "genuineness": 4.595211505889893, "identity": 0.6613330245018005, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/c4bf6cb581516889ef8c7c1b.mp3" }, { "key": "original-vn_VOLT_reg_low-es-1-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "1cbde492ec8cbc99f0f706c33b72b5b063fc083a2e5bcd51c326708231fd37c8", "task_id": "vn_VOLT_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.086154937744141, "genuineness": 4.3504462242126465, "identity": 0.5680684447288513, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/es/48af88bf79f5680fdd71b6bd.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "1e3a762c6d552d238be6fd80cee902cdb0958d0337cbaf9145cfee518674de4f", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646074056625366, "target_z": 1.749999982841793, "ce": 6.284721851348877, "genuineness": 4.2374773025512695, "identity": 0.8950198888778687, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/ac4d8ee4b6e36d80ecb3e9b1.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "28cb94ebe7ebd113b3b362031b66d85afff3ecfce096a55e2c59c41897237b5e", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.149696350097656, "genuineness": 4.055472373962402, "identity": 0.6712073087692261, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/0c8dd65cc036e33f9adaedf9.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "5db6d462d3da8021f7b88e700f774d66ae38386afeaf05825a1c3da5b12e9560", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1238985061645508, "target_z": 1.79687496890075, "ce": 6.157639026641846, "genuineness": 4.15588903427124, "identity": 0.6814002394676208, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/f897a3f303644b7121330fae.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "f98ec160c89a0df38790d471be76e366deb5dd3b3134556b9d1bea44b190025a", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1510378122329712, "target_z": 1.76562493243956, "ce": 6.070269584655762, "genuineness": 4.168440818786621, "identity": 0.7942636013031006, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/95fddeaf7146f59632c7e8ac.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "60c37d2b268d9e5264b3705ae1e655084782aaf59a129d23a18b30e251e2f293", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.23706579208374, "genuineness": 4.7081804275512695, "identity": 0.75860595703125, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/f65b6fddba47e2d3bf91b0d7.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "5c284eac816a5f04e522d865ecaf3fd6bf1e14d2db2a9b130bc9760a62d8b225", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0899744033813477, "target_z": 1.8359374801608233, "ce": 5.935244083404541, "genuineness": 4.250029563903809, "identity": 0.598817765712738, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/3c7c9523214273aa08e64e18.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "79f620ff78c85592db354648187d694a7ef8a8afbdcb3992d7cd7c4012866aed", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.042480707168579, "target_z": 1.8906249410186635, "ce": 6.133810997009277, "genuineness": 4.281409740447998, "identity": 0.7495362758636475, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/7ecb0d31b79b1cf90c76047d.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-0-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "882170c4facfdc4764c6def49f8a1e0ba908c01de799a5645d28a4ff1ac73f83", "task_id": "vn_VOLT_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0967592000961304, "target_z": 1.8281250053619398, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.6863895058631897, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/17d136479086d3626cf75573.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "e3276a4192ba5a612b233c9790cd1f88107e654471b0c4ce1126d445cbec6b84", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.32744300365448, "target_z": 1.5625000386059658, "ce": 6.125868320465088, "genuineness": 4.419482707977295, "identity": 0.7530368566513062, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/596613878d738a14073fa718.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "ab555f6b86fad98b01d6ceed3bfa06a730fbdc926016c67b784da4d719c6a3eb", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3545823097229004, "target_z": 1.531250002144776, "ce": 6.300607204437256, "genuineness": 4.570107460021973, "identity": 0.7461991906166077, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/9486cff23f8b3909bd0773b3.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "53c8315ea44f08bf6811767fb26fb7afef9538825469d8066297e98d0475ab92", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4699242115020752, "target_z": 1.3984375187667888, "ce": 6.165581703186035, "genuineness": 4.375550270080566, "identity": 0.7101370096206665, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/09727e3a5b15aa2977490351.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "4d032a61504d0018c9365cb133f4acab57bc889d546cd5a776ebf2cb48fab090", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3138734102249146, "target_z": 1.5781249882037327, "ce": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.8621250987052917, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/2d43c927137004cef632016b.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "24ee25644d86737d92fa2c403f4f79dd6407619430141a6feca9162713ec7f79", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.374936819076538, "target_z": 1.5078124404824695, "ce": 6.165581703186035, "genuineness": 4.20609712600708, "identity": 0.7647276520729065, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/c2c22a44ab7e557c232c9c7e.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "ddf2f0c80984425810dedf3add14c47555a9f5b248c6697ff8356c281d25f945", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.083189606666565, "target_z": 1.8437499549597067, "ce": 6.046441555023193, "genuineness": 4.770940780639648, "identity": 0.6920686960220337, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/28d5848343fb7e18029fa802.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "a9cc87161981c453d833472ebea5f02e1d2560994bcda444e3e1e83916474f92", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1103287935256958, "target_z": 1.8125000557641726, "ce": 6.1894097328186035, "genuineness": 4.6956281661987305, "identity": 0.8658370971679688, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/f38b1cd200e95a8272ad4da0.mp3" }, { "key": "original-vn_VOLT_reg_low-fr-1-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "1df5f69281baad936cb9e8fec29623d350d3a0fb5a8f3463cda1f383f27c9b26", "task_id": "vn_VOLT_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2053163051605225, "target_z": 1.7031249967828361, "ce": 6.117925643920898, "genuineness": 4.68307638168335, "identity": 0.7000628113746643, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/fr/b9d317ef3e4d8f86a3dfe777.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "e2e60e2bd41d6b6b71315fa6a86d951cf9df47638662a09f3f34f6f5e2f90e7d", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.7888107299804688, "target_z": 1.031249967828362, "ce": 6.062326908111572, "genuineness": 4.419482707977295, "identity": 0.7002941370010376, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/e7bf174ba58b5acd01997f30.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "106d8bc8550b04dc29e8558f9d99020d4fff4dce3d9f9cba34490ecf2d305a3d", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.2731645107269287, "target_z": 1.6249999742626895, "ce": 5.982900142669678, "genuineness": 4.21237325668335, "identity": 0.6109359860420227, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/bbcb7fd1fd8b288f13c69761.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c3f55f86ec5672e3474a191e644223913a7b34fdd05c3341e212508e385b0649", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.3477975130081177, "target_z": 1.5390624769436594, "ce": 5.927301406860352, "genuineness": 4.162164688110352, "identity": 0.6456661224365234, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/96266a07ba993b91df1052d8.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "5b53c7d65e94a823b17bd29b2ad3daf978c51a7f9e06406ba6a1121cffc34d69", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.415645718574524, "target_z": 1.4609374544235127, "ce": 6.046441555023193, "genuineness": 4.344170093536377, "identity": 0.6071611642837524, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/bc5b032ebfe9266d77f14437.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "f0f9faa4a4d9173a118a150ef88caa32e28062a6ed1243cac179a940d0ef7c64", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.3952912092208862, "target_z": 1.4843750160858191, "ce": 6.1894097328186035, "genuineness": 4.306513786315918, "identity": 0.8007236123085022, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/0ff808e5dd9a1bfcfdf77fe8.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "bda1fd70dc9ce7b28efbafcfcb182ff1a797952cc81e9d76db47601fc0fdf3cf", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.707392930984497, "target_z": 1.1249999399462756, "ce": 6.038498878479004, "genuineness": 4.3127899169921875, "identity": 0.7330416440963745, "hypothesis": "De deur staat er 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_VOLT_reg_low/nl/95475c2e26f03a161805a456.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "962ffc4e2be7f73655ac922f08321a2170d7d6b4aedfa5f5ee78b0688177e7a0", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.8159499168395996, "target_z": 1.0000000686328279, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.8268503546714783, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/13533a16a7d6345d000a8e1c.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-0-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "7fdff3502d3684e83ca8254a9939b8fa8907018989701be07c4e2a7596d5ee7c", "task_id": "vn_VOLT_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.2324556112289429, "target_z": 1.6718749603216463, "ce": 6.221180438995361, "genuineness": 4.400654315948486, "identity": 0.7956579327583313, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/3f10948b211494a9a43a0907.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1701-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "e47ef81001270bf723b60e790409604266121c466c8ffddd77544ebfdf957afd", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.218885898590088, "target_z": 1.6875000471850692, "ce": 6.117925643920898, "genuineness": 4.093128204345703, "identity": 0.5926392674446106, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/bf93a6aa7212440c04fe517f.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1701-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c29ecc7a2ee2d0189e33e94b72d3ea679d8104a1c89324b4fd8ee11e7a9cc3da", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.191746711730957, "target_z": 1.7187499463806033, "ce": 6.268836498260498, "genuineness": 4.287685871124268, "identity": 0.7136563062667847, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/0859fe4c56a7f39937215dc4.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1701-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "b7b8b0f2ca24ed009a17447914a3375434dd4ab126253e41de1c68af17b4932b", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.5581269264221191, "target_z": 1.296874934584336, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.5428715944290161, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/a2778b8e0614b88fdc6a15b0.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1701-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "aba81e8da179467d869e5dc9c61a9e72b05f8a5d6ec0540c9a069f0c091cda85", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.178176999092102, "target_z": 1.734375033244026, "ce": 6.173524379730225, "genuineness": 4.193545341491699, "identity": 0.7635843753814697, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/e8b4f7a5386e7fa7d437517c.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1702-scale0.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "a8624f85f184d94f9ec60a8b4740c0aa118b14df8732cdd6618c07aff3183392", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.5513420104980469, "target_z": 1.3046875466488752, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.7327127456665039, "hypothesis": "Ik heb het kleine doosje gevonden en 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/audio/vn_VOLT_reg_low/nl/45a64d4e674c1d1bd9567de3.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1702-scale0.5", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "c95f2d38f50f4cb264cb928e81ae77dbda26d93d44935e4723a2ae63436a110a", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.3003038167953491, "target_z": 1.5937499378014996, "ce": 6.1020402908325195, "genuineness": 4.20609712600708, "identity": 0.5291665196418762, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/eca9fa3babb5a79cd296b628.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1702-scale0.75", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "1d4a3a27697d36f26128bb95d484c0245605009255ccad8c41a4720afdcf3556", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 0.9814172983169556, "target_z": 1.9609374887399267, "ce": 6.006728172302246, "genuineness": 4.306513786315918, "identity": 0.6926670074462891, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/0bdd5924cb6e27730209c45a.mp3" }, { "key": "original-vn_VOLT_reg_low-nl-1-seed1702-scale1.0", "bucket": "vn_VOLT_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": [ "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "ec0579decfd6a8196614ca2e6d3f476c20c6ced7cba790a8f8cf886e74e76e73", "mp3_sha256": "d8379f4c709445029e4ca33525aa873047c173025d9d81e67479fba827733e27", "task_id": "vn_VOLT_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Emotionally steady, stable delivery without abrupt changes. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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(Emotionally steady, stable delivery without abrupt changes. 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": 1.2595947980880737, "target_z": 1.6406250611261124, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.510790228843689, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_VOLT_reg_low/nl/1b8aab1cb0e94c27122c12a4.mp3" } ], "source_standard_audio_unchanged": true }