{ "complete": true, "report": { "bucket": "vn_WARM_reg_high", "label": "very warm vocal delivery", "definition": { "id": "vn_WARM_reg_high", "category": "bipolar", "head": "vn_WARM_reg", "pole": "high", "label": "very warm vocal delivery", "floor": 4.0, "canonical_dimension_name": "Warmth" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P05", "general": "Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.", "inline": [ "Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.", "Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.", "Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 1.5832030998862785, "target_raw": 4.037848806381225, "ce": 6.142448687553406, "genuineness": 4.260071122646332, "identity": 0.6118041940033436, "wer": 0.049440574475404196, "valid_fraction": 0.9, "identity_delta": 0.018058379739522935, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.630468803521046, "target_raw": 4.0806174039840695, "ce": 6.036314749717713, "genuineness": 4.226494312286377, "identity": 0.5878281921148301, "wer": 0.052957731338149294, "valid_fraction": 0.9, "identity_delta": -0.0059176221489906315, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.5832030998862785, "target_raw": 4.037848806381225, "ce": 6.142448687553406, "genuineness": 4.260071122646332, "identity": 0.6118041940033436, "wer": 0.049440574475404196, "valid_fraction": 0.9, "identity_delta": 0.018058379739522935, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.547265637762711, "target_raw": 4.0053306221961975, "ce": 6.120507073402405, "genuineness": 4.213942241668701, "identity": 0.5758510857820511, "wer": 0.02330237358101135, "valid_fraction": 0.95, "identity_delta": -0.01789472848176956, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.4132812952870388, "target_raw": 3.884094250202179, "ce": 6.217407608032227, "genuineness": 4.260384953022003, "identity": 0.6135930463671684, "wer": 0.009191176470588236, "valid_fraction": 1, "identity_delta": 0.019847232103347778, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.373046854003282, "target_raw": 3.8476879239082336, "ce": 6.21482629776001, "genuineness": 4.279840648174286, "identity": 0.5816934153437614, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.012052398920059205, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.3707030998862784, "target_raw": 3.8455671668052673, "ce": 6.200529432296753, "genuineness": 4.313417446613312, "identity": 0.6021408274769783, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": 0.008395013213157655, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.2699218704712962, "target_raw": 3.75437479019165, "ce": 6.201323747634888, "genuineness": 4.302748203277588, "identity": 0.5547474734485149, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.03899834081530571, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.1865234731120804, "target_raw": 3.6789113521575927, "ce": 6.20092658996582, "genuineness": 4.276075053215027, "identity": 0.5826042920351029, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.011141522228717804, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.1718750428168367, "target_raw": 3.6656566500663756, "ce": 6.181467056274414, "genuineness": 4.2450086832046505, "identity": 0.5652974262833595, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.02844838798046112, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.630468803521046, "target_raw": 4.0806174039840695, "ce": 6.036314749717713, "genuineness": 4.226494312286377, "identity": 0.5878281921148301, "wer": 0.052957731338149294, "valid_fraction": 0.9, "identity_delta": -0.0059176221489906315, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.5839843293012608, "target_raw": 4.038555705547333, "ce": 6.132619619369507, "genuineness": 4.260384964942932, "identity": 0.6052298158407211, "wer": 0.02907937736498108, "valid_fraction": 1, "identity_delta": 0.011484001576900483, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.5355468671776933, "target_raw": 3.994726836681366, "ce": 6.049221563339233, "genuineness": 4.234025621414185, "identity": 0.5802646458148957, "wer": 0.03930641554867561, "valid_fraction": 0.9, "identity_delta": -0.013481168448925019, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.491015618001094, "target_raw": 3.95443252325058, "ce": 6.144732189178467, "genuineness": 4.2525399088859555, "identity": 0.6212875008583069, "wer": 0.04175417096663227, "valid_fraction": 0.9, "identity_delta": 0.027541686594486237, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.4662108759507972, "target_raw": 3.931987833976746, "ce": 6.076028108596802, "genuineness": 4.182248294353485, "identity": 0.5603329069912434, "wer": 0.03884029927760578, "valid_fraction": 0.95, "identity_delta": -0.033412907272577286, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.4636718507096793, "target_raw": 3.929690384864807, "ce": 6.128449749946594, "genuineness": 4.269171345233917, "identity": 0.6132439777255059, "wer": 0.03267737358101135, "valid_fraction": 0.95, "identity_delta": 0.019498163461685182, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.4527343490628777, "target_raw": 3.9197935342788695, "ce": 6.168758773803711, "genuineness": 4.271681809425354, "identity": 0.5961892232298851, "wer": 0.015490196078431372, "valid_fraction": 1, "identity_delta": 0.002443408966064453, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.3851562458829965, "target_raw": 3.858645164966583, "ce": 6.231108713150024, "genuineness": 4.290823709964752, "identity": 0.5790349110960961, "wer": 0.011721491228070174, "valid_fraction": 1, "identity_delta": -0.01471090316772461, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.3535156706987392, "target_raw": 3.83001503944397, "ce": 6.196955251693725, "genuineness": 4.283292472362518, "identity": 0.5859199091792107, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.007825905084609985, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.3503906212946968, "target_raw": 3.8271873235702514, "ce": 6.160816073417664, "genuineness": 4.162792503833771, "identity": 0.5487287983298301, "wer": 0.009052287581699346, "valid_fraction": 1, "identity_delta": -0.045017015933990476, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.3417968474160764, "target_raw": 3.819411206245422, "ce": 6.0249964714050295, "genuineness": 4.183817207813263, "identity": 0.5907619148492813, "wer": 0.05645833333333333, "valid_fraction": 0.95, "identity_delta": -0.0029838994145393372, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.3044921712378361, "target_raw": 3.7856558918952943, "ce": 6.202912282943726, "genuineness": 4.280468225479126, "identity": 0.5505218848586082, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0432239294052124, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.289843727768181, "target_raw": 3.772401177883148, "ce": 6.171141600608825, "genuineness": 4.239987778663635, "identity": 0.6166273221373558, "wer": 0.008350533195734434, "valid_fraction": 1, "identity_delta": 0.022881507873535156, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 1.2701172074674671, "target_raw": 3.7545515418052675, "ce": 6.171141576766968, "genuineness": 4.242498207092285, "identity": 0.5532683365046978, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": -0.040477477759122846, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.269921857296885, "target_raw": 3.7543747782707215, "ce": 6.207280707359314, "genuineness": 4.267916107177735, "identity": 0.5491080619394779, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.04463775232434273, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.2251952711241145, "target_raw": 3.713903713226318, "ce": 6.158036160469055, "genuineness": 4.2023315787315365, "identity": 0.6012680381536484, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.007522223889827729, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.1787109813460876, "target_raw": 3.6718421816825866, "ce": 6.232697248458862, "genuineness": 4.246577632427216, "identity": 0.5949970692396164, "wer": 0.0058823529411764705, "valid_fraction": 1, "identity_delta": 0.001251254975795746, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.1666016026407844, "target_raw": 3.660884952545166, "ce": 6.225151705741882, "genuineness": 4.193074405193329, "identity": 0.6040232494473458, "wer": 0.013859649122807016, "valid_fraction": 0.95, "identity_delta": 0.010277435183525085, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.1654297124078714, "target_raw": 3.659824562072754, "ce": 6.190601086616516, "genuineness": 4.256619322299957, "identity": 0.5782951325178146, "wer": 0.02724458204334365, "valid_fraction": 0.9, "identity_delta": -0.015450681746006011, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.15800778923893, "target_raw": 3.65310879945755, "ce": 6.194969606399536, "genuineness": 4.228063344955444, "identity": 0.6101733453571796, "wer": 0.015935672514619883, "valid_fraction": 0.95, "identity_delta": 0.016427531093358993, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.1457030998862785, "target_raw": 3.6419748425483705, "ce": 6.220783257484436, "genuineness": 4.214256072044373, "identity": 0.5831899996846914, "wer": 0.02222222222222222, "valid_fraction": 0.95, "identity_delta": -0.010555814579129219, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.567578129528704, "target_raw": 4.023710477352142, "ce": 6.100451803207397, "genuineness": 4.2836061954498295, "identity": 0.5773666486144066, "wer": 0.06088020295837633, "valid_fraction": 0.85, "identity_delta": -0.016379165649414062, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.5660156838731505, "target_raw": 4.022296690940857, "ce": 5.931272959709167, "genuineness": 4.201704001426696, "identity": 0.5658045466989279, "wer": 0.0967640608875129, "valid_fraction": 0.85, "identity_delta": -0.02794126756489277, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.2048828057069443, "target_raw": 3.6955238819122314, "ce": 6.200926566123963, "genuineness": 4.277644002437592, "identity": 0.529593002051115, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0641528122127056, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.1853515433559336, "target_raw": 3.6778509259223937, "ce": 6.198940944671631, "genuineness": 4.253167462348938, "identity": 0.5937458142638207, "wer": 0.0061111111111111106, "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": 1.143945317234554, "target_raw": 3.640384304523468, "ce": 6.210457801818848, "genuineness": 4.285489094257355, "identity": 0.52261912971735, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.07112668454647064, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.241406252470202, "target_raw": 3.7285722970962523, "ce": 6.177694272994995, "genuineness": 4.297413575649261, "identity": 0.576803269982338, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.016942544281482695, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.2025390120667068, "target_raw": 3.693403089046478, "ce": 6.1850412607192995, "genuineness": 4.2556779146194454, "identity": 0.5503812812268734, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.04336453303694725, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.0912500092220878, "target_raw": 3.592702703475952, "ce": 6.265659379959106, "genuineness": 4.3132919454574585, "identity": 0.5905555510520935, "wer": 0.006680337206652996, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.4284375253607418, "target_raw": 3.897808437347412, "ce": 6.231664724349976, "genuineness": 4.322078394889831, "identity": 0.6065105706453323, "wer": 0.02362417902463669, "valid_fraction": 0.96, "identity_delta": 0.01595501959323883, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.4265624588299648, "target_raw": 3.89611177444458, "ce": 6.2015620136260985, "genuineness": 4.27011281490326, "identity": 0.5929865968227387, "wer": 0.02902552824749621, "valid_fraction": 0.94, "identity_delta": 0.0024310457706451416, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.4284375253607418, "target_raw": 3.897808437347412, "ce": 6.231664724349976, "genuineness": 4.322078394889831, "identity": 0.6065105706453323, "wer": 0.02362417902463669, "valid_fraction": 0.96, "identity_delta": 0.01595501959323883, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": -0.0018750665307769884, "ci95_z": [ -0.05828137696838877, 0.05234372776818093 ], "language_deltas": { "en": 0.037499907779121024, "de": 0.0890625247020211, "fr": -0.09531264985892839, "nl": -0.04296884304427967, "es": 0.0023437277681809922 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.030102710723876953, "delta_genuineness": -0.051965579986572266, "delta_wer": 0.00540134922285952, "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": -0.0018750665307769884, "ci95_z": [ -0.05828137696838877, 0.05234372776818093 ], "language_deltas": { "en": 0.037499907779121024, "de": 0.0890625247020211, "fr": -0.09531264985892839, "nl": -0.04296884304427967, "es": 0.0023437277681809922 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.030102710723876953, "delta_genuineness": -0.051965579986572266, "delta_wer": 0.00540134922285952, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.33718751613865383, "ci95_z": [ 0.29726559823947707, 0.3724218994550009 ], "language_deltas": { "en": 0.32851567069873916, "de": 0.2664061734237344, "fr": 0.33515631175505295, "nl": 0.358593800227443, "es": 0.3972656245882996 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.03399465560913086, "delta_genuineness": 0.008786449432373047, "delta_wer": 0.016943841817983692, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 5.0, "reason": "Use manual-only configuration; no clear qualifying incremental adapter gain" }, "reference_only": true, "updated": 1793036131.7224436, "example": { "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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.", "language": "fr", "seed": 1702, "scale": 0.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "009d9fc54498870ca5f8c0b8d057c8e4eac380b9b53b0b22284effee64211285", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a3902e0ce8e03e9576f8dee897f8758fa43892e0a0dbb3dce95ae25fdd7e6f93", "generation_seconds_batch": 24.897501318948343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344297409057617, "target_z": 1.921874687107732, "ce": 5.919358730316162, "genuineness": 4.337893962860107, "identity": 0.5592552423477173, "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_WARM_reg_high/009d9fc54498870ca5f8c0b8d057c8e4eac380b9b53b0b22284effee64211285.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "00a314c746b3d9b926ff93ad5eb4cb1d6edebc039c9ca1509176462dba30d696", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6865af3f407e387752c2f3ea65c8c61e722b55f992e8efa96184bc28822117de", "generation_seconds_batch": 18.52385436394252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.308549880981445, "genuineness": 4.055472373962402, "identity": 0.3996659517288208, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/00a314c746b3d9b926ff93ad5eb4cb1d6edebc039c9ca1509176462dba30d696.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7cafe34adfebb19d56634c1d10731c1ca03baf5a936c357e0555efef8395a815", "generation_seconds_batch": 21.073604903067462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.6425172686576843, "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_WARM_reg_high/00aa858e5dae243ca685e47cab6692a7a6eadbedd8b1d63fcfb174f2b3a210fb.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "00b4c89053a47ac06b5e27d27da15da3aa2dbf4cc1ebef670c0d5d0d89ae0091", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "280c7cf30200cea3e07b00f32287a58ba8da47076d20571921a9bbb95f8cea84", "generation_seconds_batch": 16.057800026028417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.062326908111572, "genuineness": 3.9550557136535645, "identity": 0.6303943991661072, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/00b4c89053a47ac06b5e27d27da15da3aa2dbf4cc1ebef670c0d5d0d89ae0091.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "01173e5ecf5b8c328710b549cc10f8a2aa75084723b5b41b2d833db9be2043d0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "73a916097082bb87e2efbba1d3007859b3356bfe3e3a3555409dc705f1bb48cf", "generation_seconds_batch": 25.469032046967186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.5819617509841919, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/01173e5ecf5b8c328710b549cc10f8a2aa75084723b5b41b2d833db9be2043d0.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "0145d766dc0e3ff0cc311dd434b40139380d05d100f80ce9e5b2dcc4def2a935", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "97f94c2e14c2f3ede02d9ac4590da9d8633258efd5b03ef8be79571f406bce7d", "generation_seconds_batch": 24.874132041004486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.348263263702393, "genuineness": 4.557555198669434, "identity": 0.7423170804977417, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0145d766dc0e3ff0cc311dd434b40139380d05d100f80ce9e5b2dcc4def2a935.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "0148fbfa68c178b8b1dd1f7683100723151e9aa632a5cbeb646f8a7153d77bc8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "147430e561e6205f4090eb1c278b804c1b7a3cb0d9597738f8461a80cefc6e79", "generation_seconds_batch": 21.507822999963537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.078212261199951, "genuineness": 4.381826400756836, "identity": 0.7612346410751343, "hypothesis": "De deur 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_WARM_reg_high/0148fbfa68c178b8b1dd1f7683100723151e9aa632a5cbeb646f8a7153d77bc8.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "01635d9f8f91a06d3bebd4a55132b79f4ba006ef49b31e7c16835b5d2ac6dbb2", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "98e10450d63cd4d703a652ed21a43b5946162fd613002c6174ce17d731e0ad29", "generation_seconds_batch": 17.002740437048487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3758203983306885, "target_z": 0.8515625905740776, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.838520348072052, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/01635d9f8f91a06d3bebd4a55132b79f4ba006ef49b31e7c16835b5d2ac6dbb2.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "016fbb3fc8298cc84aa2b14e75d8fe3bb47ec73cfd6bcd1424ce6d0003dd5a02", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8e95fffd81be9516ff403d2084f37fc307c13609568ba61ac6fa138899f005b7", "generation_seconds_batch": 18.456865523941815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457404613494873, "target_z": 2.0468750823400708, "ce": 6.157639026641846, "genuineness": 4.6203155517578125, "identity": 0.6872212886810303, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/016fbb3fc8298cc84aa2b14e75d8fe3bb47ec73cfd6bcd1424ce6d0003dd5a02.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "01c1759c13955a802d158871dd3797defb3a7e2c2fb0176a5282b9e35ee8aecc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f93f47e6abf2fd474c56e494f0e5d9cfa4b805e12f9cac06ba8e31ddfbeac472", "generation_seconds_batch": 23.376218984019943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.040323257446289, "target_z": 1.5859377387862046, "ce": 6.403861999511719, "genuineness": 4.293961524963379, "identity": 0.7009012699127197, "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_WARM_reg_high/01c1759c13955a802d158871dd3797defb3a7e2c2fb0176a5282b9e35ee8aecc.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c01a11d1e769e24fcd6a26b4b96225b5114ec8ed338c078bf37cb42899a846ce", "generation_seconds_batch": 33.26402025995776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.499174118041992, "genuineness": 3.9927117824554443, "identity": 0.2892538905143738, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very warm vocal delivery. 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": "a08f3c99b5c9d6ba078d0a1833eeae3562ee8fc037943ee7ec9d26ba146d4ccc", "generation_seconds_batch": 15.778318897006102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.09409761428833, "genuineness": 4.432034492492676, "identity": 0.6629713773727417, "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_WARM_reg_high/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "026b9ee5681d0d1fcea53259fa12b3b835af1a61d6ebb6c3c24d8974273cd93d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "0e2ddfd07a2553f6b31106573bb599702ef906a8d78c6216b9433776516757f6", "generation_seconds_batch": 23.47902188799344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 5.824047088623047, "genuineness": 4.344170093536377, "identity": 0.7910773754119873, "hypothesis": "J'ai trouvé la Mimi de Kitt. J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle fait.", "wer": 0.3684210526315789, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/026b9ee5681d0d1fcea53259fa12b3b835af1a61d6ebb6c3c24d8974273cd93d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "027922cb6bc69284b693f491891a820d3e12c613a908ac82a159c0f1e9cb77e2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c25ff473cb606cc137f47f0a46ee5802b65ba2cb8d2791519999e9ca283be3fb", "generation_seconds_batch": 24.805956483003683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.141753673553467, "genuineness": 4.494794845581055, "identity": 0.477864146232605, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/027922cb6bc69284b693f491891a820d3e12c613a908ac82a159c0f1e9cb77e2.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "029664d0abb942bf3c6be65f57f62513af18b31de00baed4eb51a54506a67114", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e160ab87308fe23f1c3c86be43367b6daa24f40b36f0e4bb396e99264e32b6ca", "generation_seconds_batch": 16.13402673904784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.395919322967529, "genuineness": 4.632867813110352, "identity": 0.4567544460296631, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/029664d0abb942bf3c6be65f57f62513af18b31de00baed4eb51a54506a67114.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "02a8158f858132f04394651d23076e9fab784e2cb16e110df29ea8f84700a5ae", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "07518c377404ad5039d681c852b3e16fff315eb7cf99e0306763f03e3acb776d", "generation_seconds_batch": 16.901269335998222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.213237762451172, "genuineness": 4.005263805389404, "identity": 0.32898616790771484, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/02a8158f858132f04394651d23076e9fab784e2cb16e110df29ea8f84700a5ae.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "02ac104901b24152bff3f57171a0cb38d341ff697d5bdbec92a800d631be76ba", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "66292c59bb492be5cd59bdd7de6e61af60975968db9ebf2d73e70c33dceecfb8", "generation_seconds_batch": 16.24828257295303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.372574329376221, "target_z": 1.9531249176599295, "ce": 6.046441555023193, "genuineness": 4.413206577301025, "identity": 0.6538240313529968, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/02ac104901b24152bff3f57171a0cb38d341ff697d5bdbec92a800d631be76ba.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "03219de802a9a2903320bdac53368aa3b46da07c7ad9445ac91ad487ad944285", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d9756c637b2ff2955c2cfbe99f52f85ceab61c371bc8c9dd283c6cffc4d0e640", "generation_seconds_batch": 15.588876973022707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.3641486167907715, "genuineness": 3.936227321624756, "identity": 0.6731619834899902, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/03219de802a9a2903320bdac53368aa3b46da07c7ad9445ac91ad487ad944285.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "03410e15891735a37dc52d415d3c7396c616dad169eb563b4f4302f5679c0703", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5749324349b4dc612cd6b96c8fc30940ac5fbd03ef3e9e649ee62ef2b8d6e8d1", "generation_seconds_batch": 22.77544435893651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.139291763305664, "target_z": 1.6953127552542187, "ce": 6.006728172302246, "genuineness": 4.65797233581543, "identity": 0.6813617944717407, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/03410e15891735a37dc52d415d3c7396c616dad169eb563b4f4302f5679c0703.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "034b721518568abc6aaafd3d54004f1d01082dc879f1f3367cf6f9116994a467", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b17f3b708bf2ebaca392c932eb85389b7a7b87f8fa0192e82fcc38e5324a3b35", "generation_seconds_batch": 22.061700516031124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0120463371276855, "target_z": 1.554687508234007, "ce": 6.24500846862793, "genuineness": 4.055472373962402, "identity": 0.7048419117927551, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/034b721518568abc6aaafd3d54004f1d01082dc879f1f3367cf6f9116994a467.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "038797a51c038ce3cf385f4c3b7f2741a14ce51240f0b82b004a15965deb7f64", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "45f03d141b236ce5ea5fb632ba0b5d5f6df30915438f67a809d404d01751066c", "generation_seconds_batch": 23.184207169921137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.139002799987793, "target_z": 0.589843688244947, "ce": 6.1894097328186035, "genuineness": 4.2249250411987305, "identity": 0.6667091250419617, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/038797a51c038ce3cf385f4c3b7f2741a14ce51240f0b82b004a15965deb7f64.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "039304b2e5836285155863082e8fc7a02ee369329c918ff8f88d785343640f14", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "066c9b6263e0cc5fc24aff5d841aae4f82bfc85cdc5bf2268bf8199e37181ed5", "generation_seconds_batch": 22.061112171038985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.173524379730225, "genuineness": 4.325342178344727, "identity": 0.3232918381690979, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/039304b2e5836285155863082e8fc7a02ee369329c918ff8f88d785343640f14.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "039506c91ecd2c92baa2952204304b37c8eea548b059a507b49e323006a98fcf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6f4b74e16c31c639b0f11fdc1807c405b85298a48d0e0f1a1aa7ba88e57bed00", "generation_seconds_batch": 24.070775900036097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.300607204437256, "genuineness": 4.15588903427124, "identity": 0.6958044767379761, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/039506c91ecd2c92baa2952204304b37c8eea548b059a507b49e323006a98fcf.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "03d9e88f1d59291c813de61f144716dc78ea00945f699592641cc20ebea9b817", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5054ce532ab8a5630c96b264221d05bc4a8bda8b7f388b2ede918e8f28c86241", "generation_seconds_batch": 23.205246609984897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.09409761428833, "genuineness": 3.9173994064331055, "identity": 0.5498319864273071, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/03d9e88f1d59291c813de61f144716dc78ea00945f699592641cc20ebea9b817.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "041078de4577c2d420c329fa8789771ae09cadf4ea50ed0c5b137c674824683b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "dfb46e9c4fe9bf19a1036637532fffe091998c9c959c8f7db2f99e51916c3453", "generation_seconds_batch": 23.613387380028144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.118083953857422, "target_z": 1.6718749505959576, "ce": 6.23706579208374, "genuineness": 4.243753433227539, "identity": 0.5788804292678833, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/041078de4577c2d420c329fa8789771ae09cadf4ea50ed0c5b137c674824683b.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "046c94acc07214d6d622f9ef997a1faf27621d98e7dd3eb6e8210cf34654218e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ca0cb4bbad2bca9f95bd993246a7ec6669c8fe76e4b76b65fd81ab4385a1a8ab", "generation_seconds_batch": 23.352548515074886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.6045130491256714, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/046c94acc07214d6d622f9ef997a1faf27621d98e7dd3eb6e8210cf34654218e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "04ce055e10d44db02b225b9a308c30acf67f08d6a4d9035d3dc4f34e8677b4a2", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "57303969774bc6fd1cd7d7fdbc9857f50729c638aeeaf3738e6116266d67700c", "generation_seconds_batch": 15.457227946957573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.252951145172119, "genuineness": 4.293961524963379, "identity": 0.7212152481079102, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/04ce055e10d44db02b225b9a308c30acf67f08d6a4d9035d3dc4f34e8677b4a2.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "05750d07795498a1166c02fbd01b49cf2bec593220f3fd45990afd30ceb46e2e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "054610c4c50fe4e1100c0f89d35e72fed9f113d2c280256d649573979693b822", "generation_seconds_batch": 15.206619591917843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.403861999511719, "genuineness": 4.457139015197754, "identity": 0.4433839023113251, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/05750d07795498a1166c02fbd01b49cf2bec593220f3fd45990afd30ceb46e2e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d0134ffa6adc64d1ae9186d3b5e654003abf70d1c13ae36289d63eb752752a21", "generation_seconds_batch": 21.213882546988316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.356205940246582, "genuineness": 4.306513786315918, "identity": 0.49587196111679077, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very warm vocal delivery. 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": "cb04371349f6b3c17245db0ec20ea0acc5d1b5503154899662406f9a6c7ae5b7", "generation_seconds_batch": 15.431697850930505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.348263263702393, "genuineness": 4.469690799713135, "identity": 0.6327304244041443, "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_WARM_reg_high/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "05a37a0cc8da292dc2cb266c6a6757769ddec3fb22ae04692c0b8e6838c415ea", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "27766882158be73cc7fa1f489b2f8dd90e47d1e85c60f871882c356ad4b8a05c", "generation_seconds_batch": 24.43682372092735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.308549880981445, "genuineness": 4.124508857727051, "identity": 0.7476171255111694, "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_WARM_reg_high/05a37a0cc8da292dc2cb266c6a6757769ddec3fb22ae04692c0b8e6838c415ea.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "05d30d4b2a1b04ac6731e34ba21f3f59d4651855cf215d814fba9e4cb09a74d0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 130, "seconds_generated": 10.4, "mp3_sha256": "bc7e96e681f6fff66504cb11f3a196a01f7c39d71bcf9bfa4bb9271a91704bf6", "generation_seconds_batch": 27.07980597706046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.348263263702393, "genuineness": 4.080576419830322, "identity": 0.47324782609939575, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/05d30d4b2a1b04ac6731e34ba21f3f59d4651855cf215d814fba9e4cb09a74d0.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "06150e899e781c2ebc49042506ece79ae39f64d2d1badf74bf1c4423066eb974", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e01c29679c914756a46006762406a8ce27d8170bee3596e1d04bd17619ba0ed3", "generation_seconds_batch": 16.345727602019906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.799971103668213, "target_z": 1.320312623510106, "ce": 6.292664527893066, "genuineness": 4.218649387359619, "identity": 0.6032503247261047, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/06150e899e781c2ebc49042506ece79ae39f64d2d1badf74bf1c4423066eb974.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "0677ef472d48704754f20dbbc96fe8babda2a71d07a1025f7eb72d3b533389ad", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4b4388f0bf84aee2472f92cf09138dfffb0b43944501e39f493c26a673c582ae", "generation_seconds_batch": 17.405034702969715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.188776016235352, "target_z": 1.7500002634882257, "ce": 6.252951145172119, "genuineness": 4.532451152801514, "identity": 0.7152919769287109, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0677ef472d48704754f20dbbc96fe8babda2a71d07a1025f7eb72d3b533389ad.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "06a2a47fb8336862bbab47c90e666e2f2421816b15b17953c2020e29e56a5eac", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "70de7721cc294841b7d5787d5f09f9cc20f3be905b1985ef1d2a0b3259b149ff", "generation_seconds_batch": 22.09716922591906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4217700958251953, "target_z": 0.9023438858611164, "ce": 5.895530700683594, "genuineness": 4.394378662109375, "identity": 0.7451637387275696, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/06a2a47fb8336862bbab47c90e666e2f2421816b15b17953c2020e29e56a5eac.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "06ff521b86b78e123656d25350603f5fd823ba609b6e5e0632194f00bb1ef3a6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "58ae22d3fe4b0f66a1fe0cea862a4dc2cdda0286a0c37a999cda9d2b2091e1e7", "generation_seconds_batch": 23.7681716469815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 6.157639026641846, "genuineness": 4.381826400756836, "identity": 0.5470078587532043, "hypothesis": "Ik 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_WARM_reg_high/06ff521b86b78e123656d25350603f5fd823ba609b6e5e0632194f00bb1ef3a6.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "079d4fc7debfe47a8c340da36e84644133bd7f8d1753d3a32f8d5ec2d262ecbe", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "49aad3b184c9c747c1dae154443eee1d112fcddd235bc6df2c614e8d5d5b4240", "generation_seconds_batch": 22.761452679056674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.443265914916992, "target_z": 2.031249703575746, "ce": 6.173524379730225, "genuineness": 4.369274139404297, "identity": 0.6858028173446655, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/079d4fc7debfe47a8c340da36e84644133bd7f8d1753d3a32f8d5ec2d262ecbe.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "07f2fd1cfd3796e9193ad3643f2f94503863e9abe478ab6c0c6eaafeb12b8da4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e16e6793fcdd9906781471f4166db1871acb39b1f7077af9fb5ce484fc1bb9cb", "generation_seconds_batch": 23.183197143953294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 6.173524379730225, "genuineness": 4.482243061065674, "identity": 0.6633709669113159, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/07f2fd1cfd3796e9193ad3643f2f94503863e9abe478ab6c0c6eaafeb12b8da4.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "082f43beb3ff1b7186b454db71d7c7a483fe88c38ef576202092c1c32a0ddc9d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "160aaec27ef3ae9e49b2cd5a6010df6a83147c32368c9890ccf56b1caefb8204", "generation_seconds_batch": 16.364639181061648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.202914237976074, "target_z": 1.7656251152760987, "ce": 6.133810997009277, "genuineness": 4.444586753845215, "identity": 0.7822728157043457, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/082f43beb3ff1b7186b454db71d7c7a483fe88c38ef576202092c1c32a0ddc9d.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "0864442becd56e4c7a791af8c31ff5fa305b913db9520d647cd89c5e431c5fd9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "aa2aec3944d7559c4a36df1ab85ba946916cd0ca6b75e96d00cc77995f0ed435", "generation_seconds_batch": 24.413496519089676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.287744522094727, "target_z": 1.85937527995624, "ce": 5.974957466125488, "genuineness": 4.494794845581055, "identity": 0.5186593532562256, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0864442becd56e4c7a791af8c31ff5fa305b913db9520d647cd89c5e431c5fd9.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "08e75ad87ee5ad337bcc7b58560975367b9e1064ee855827a7a97997ba0581a0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1872bf0e82b0c00b2b8c78735bf07cd3275ebc69d9345939245f2cccad049831", "generation_seconds_batch": 21.957208935986273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.711894989013672, "target_z": 2.328125049404042, "ce": 5.681078910827637, "genuineness": 4.344170093536377, "identity": 0.6021751761436462, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/08e75ad87ee5ad337bcc7b58560975367b9e1064ee855827a7a97997ba0581a0.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "08fbde6f9bb02ac22df9d9a67bf6a78559d0e1ff48f1e25564fcb2bfac94137e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "52d221aa09dbe2e5b95800a1d7c7b35d6164faa2cc3734faa1b429c6e72f9456", "generation_seconds_batch": 21.775823137024418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.24500846862793, "genuineness": 3.9676074981689453, "identity": 0.3072558641433716, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/08fbde6f9bb02ac22df9d9a67bf6a78559d0e1ff48f1e25564fcb2bfac94137e.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "098328cfd5dd43db225e827f4be76753e55d7dcae437210ed44acfabe8ad30b7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4c69e3c6a15c76066974f1406db7dd126f1f572497ca485ee9197bd6688da1db", "generation_seconds_batch": 18.579376109992154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 5.982900142669678, "genuineness": 3.8420867919921875, "identity": 0.5637360215187073, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/098328cfd5dd43db225e827f4be76753e55d7dcae437210ed44acfabe8ad30b7.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "eb3ec42cbbadda2bc5a6de372705ba9ebf8b16038245b67f8e9029c4b4091166", "generation_seconds_batch": 20.988031437969767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.125868320465088, "genuineness": 4.262581825256348, "identity": 0.5913751125335693, "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_WARM_reg_high/0a100968e72c95d714771dd2edf027554d30a9aa72525db0354bf7dddea152b1.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "0a138e201ba3779e99129b599530c30b9931f3630d45a6b106669c0e29706107", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7199950d4e0ea674a65c82acb8b291d77e5d3dd90421ae839cc51ccafdb07990", "generation_seconds_batch": 17.274533401010558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.141753673553467, "genuineness": 3.8671910762786865, "identity": 0.5019460916519165, "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_WARM_reg_high/0a138e201ba3779e99129b599530c30b9931f3630d45a6b106669c0e29706107.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "0a3585c141e8730d5acd4a3d928556f7d3d2b284ffc6fda2e68e0f98ee43b5b7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2d33c9640a697cab3ef552f1f2a072c9d15c8621dcae6ad78b44bf9bff736377", "generation_seconds_batch": 17.55741760507226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4323737621307373, "target_z": 0.9140625247020212, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.524901807308197, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0a3585c141e8730d5acd4a3d928556f7d3d2b284ffc6fda2e68e0f98ee43b5b7.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a56d2b2bfd16bd1d7afc59c586ec8ac0d7520b962a55cddee42ed2d80b309b65", "generation_seconds_batch": 17.73100562498439, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.133810997009277, "genuineness": 4.293961524963379, "identity": 0.6284019351005554, "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_WARM_reg_high/0a742cea65de1e30ecb4f04888c2749cd9443873212d4ebaf42a496a47f5f840.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "0a84bd213ca222c0fefe18efd6bcb06214e9462283a561cac58e5d6b8b630f75", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2cd27db2dc8e7777921ea685c32bb447f45748662fe4c174bf99e69693583cd8", "generation_seconds_batch": 25.27418491896242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.316492557525635, "genuineness": 4.231201171875, "identity": 0.2813242971897125, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0a84bd213ca222c0fefe18efd6bcb06214e9462283a561cac58e5d6b8b630f75.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "0b0b400131be8b1b99a6980948461116d0f10bf254eeb22b80f30e980fcea195", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "a3c5872ce7dbb82690b909f815ff0537bc63923dfb3a36b9c9a107606c7b07eb", "generation_seconds_batch": 24.219300976023078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.157639026641846, "genuineness": 4.325342178344727, "identity": 0.45165780186653137, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que haya.", "wer": 0.11764705882352941, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0b0b400131be8b1b99a6980948461116d0f10bf254eeb22b80f30e980fcea195.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "0b5bac49c5d80ec0e49638f2135675178832560098dd56022758dd3a78c9b066", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "86315cbf534820891b9c9d422faa644e6cf64389ab29e14b0b79cefd8d3ff8d3", "generation_seconds_batch": 22.31237850699108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 6.284721851348877, "genuineness": 4.293961524963379, "identity": 0.7901884317398071, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0b5bac49c5d80ec0e49638f2135675178832560098dd56022758dd3a78c9b066.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "0b721483763fa4e47b2963a32dcfb5b7212de51bafcb865faf31d77ff1003906", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "419c6398388f5f394be9a0bd2084ec47a5d6c3adb35305de5f43edcd9a12dd45", "generation_seconds_batch": 17.21395555092022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.435908317565918, "target_z": 0.9179687376489895, "ce": 6.1020402908325195, "genuineness": 4.306513786315918, "identity": 0.6422361135482788, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0b721483763fa4e47b2963a32dcfb5b7212de51bafcb865faf31d77ff1003906.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "0b99acd82bc79415d6b5757009987291e37b3549108155d4e148fb565ee377b6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e99a78a0ea41a94c54f3ecdb18910efb180c8d60ffb74b7001e418ceb58aae27", "generation_seconds_batch": 22.061700516031124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.153429985046387, "target_z": 1.7109376070420916, "ce": 6.149696350097656, "genuineness": 4.086852550506592, "identity": 0.7746882438659668, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0b99acd82bc79415d6b5757009987291e37b3549108155d4e148fb565ee377b6.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "0c2a762484d6ee167cd2a04e4fb17aab129421e52e4625397b4a6168a3896143", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4f0c5caa679e0d514bdaaac0727e8fb040a2b8931b24152cd08ac9dfce0a4f09", "generation_seconds_batch": 23.242546002031304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.082449436187744, "target_z": 0.5273437541170035, "ce": 6.213237762451172, "genuineness": 4.218649387359619, "identity": 0.7085903286933899, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0c2a762484d6ee167cd2a04e4fb17aab129421e52e4625397b4a6168a3896143.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "0c3a5c74b393e767b79b10a91a00ff05434442005b821a889972f748adcbc8ba", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "89e9daece06d91e02308000b5d4362c992158615fe8918eb663325c5c2c7af5d", "generation_seconds_batch": 22.961509054992348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.84238600730896, "target_z": 1.3671874423619506, "ce": 6.308549880981445, "genuineness": 4.111956596374512, "identity": 0.22175970673561096, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0c3a5c74b393e767b79b10a91a00ff05434442005b821a889972f748adcbc8ba.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "0c45a2872f53d890dadef53f4693b7e58d18c36767945e40f700ebdb791f9d1b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "42395c0c93953f44868f7c7a2b480b3248ca9128368182b7fe6d1db3ccc7d02b", "generation_seconds_batch": 17.708350779954344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.292664527893066, "genuineness": 4.2751336097717285, "identity": 0.6356936693191528, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0c45a2872f53d890dadef53f4693b7e58d18c36767945e40f700ebdb791f9d1b.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "0c9ce88bfc0e6837571a440aef6c7b7a7c5268d323ebc602d8973214bdb606ce", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "eba5b81f37ee444df601d4cfa44238c336dcc5de506ba50ecf5162ab7adea9ce", "generation_seconds_batch": 24.670562471961603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.149696350097656, "genuineness": 4.419482707977295, "identity": 0.7108718156814575, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0c9ce88bfc0e6837571a440aef6c7b7a7c5268d323ebc602d8973214bdb606ce.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "0c9f4361976f77b927439fa512dd3683c39dae121044794a57b4f90625f02fa3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "698cc9a75890f9425c7d9205992ac85febddf6fc5362f7e91de25b086ff03c87", "generation_seconds_batch": 18.97512251092121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.146360874176025, "target_z": 1.7031251811481551, "ce": 5.911416053771973, "genuineness": 4.758388519287109, "identity": 0.8122771978378296, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0c9f4361976f77b927439fa512dd3683c39dae121044794a57b4f90625f02fa3.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very warm vocal delivery. 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": "a62f7c8effbc74358477f41b584cd3c8b9ed3843480defea0fdfb73d70315984", "generation_seconds_batch": 17.67209276801441, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6303107738494873, "target_z": 1.1328125576380494, "ce": 6.205295085906982, "genuineness": 4.3127899169921875, "identity": 0.586241602897644, "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_WARM_reg_high/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "0cbff49a5fd81bd439905fc796d00c5d4d63a596848a2782d046f7f35f1c8afa", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "719b2721007b8abe45ccaef642a421286426157636261f5d7bb37dd08fe8c405", "generation_seconds_batch": 17.723066051956266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 5.935244083404541, "genuineness": 3.9550557136535645, "identity": 0.6089335680007935, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás, Silvia.", "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_WARM_reg_high/0cbff49a5fd81bd439905fc796d00c5d4d63a596848a2782d046f7f35f1c8afa.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "0cd9c3ca2fd584341c81b07999a3daafbaacb7dc68d6a19798103c2953eb0747", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e789f27935f89ce3dd0f72abd985c9860b6e949a367d9e3ec32cf756cd72665d", "generation_seconds_batch": 23.897229030029848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.427689552307129, "genuineness": 4.413206577301025, "identity": 0.5371261835098267, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0cd9c3ca2fd584341c81b07999a3daafbaacb7dc68d6a19798103c2953eb0747.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4b0268ac0c303465ff486566f6e0f0e87ed487e376f131b823a820bfe31ce7c3", "generation_seconds_batch": 21.224965500994585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.205295085906982, "genuineness": 4.281409740447998, "identity": 0.5495926141738892, "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_WARM_reg_high/0ce1531aef9e8e8173d1e7c7292c86593f4fab6ffac8a8798138a562e85295a5.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "0cf71a883190e7727e3195f5d41a0d6783cc0f215f4eb4ccdab569a63cd2b8d2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bccff07a179d6d3eec4db99d42613fa5a16c12647ca7a6e901c754a7f827c6b3", "generation_seconds_batch": 20.86953905900009, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.467403411865234, "genuineness": 4.168440818786621, "identity": 0.6611089706420898, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0cf71a883190e7727e3195f5d41a0d6783cc0f215f4eb4ccdab569a63cd2b8d2.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "0d528c4bd0faa37c3cbb74deeb3b81f1053d5977ba486c2d84233bd55adcae77", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "de7a555d7f0633b89739245bb38f7d50e7dd7976fde357af3f277be714b8f3fa", "generation_seconds_batch": 23.441742299939506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.188776016235352, "target_z": 1.7500002634882257, "ce": 5.8875885009765625, "genuineness": 4.068024158477783, "identity": 0.59222412109375, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0d528c4bd0faa37c3cbb74deeb3b81f1053d5977ba486c2d84233bd55adcae77.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "0d569a49ddb3363de89880902538b947a4aa3d14163b85236e9d20cb2e5fb4f4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "41653125dfea788ae58b6a93ea7c58167a2bfd0acc65e1d4b2de2d9191aa3965", "generation_seconds_batch": 23.38369412207976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.348263263702393, "genuineness": 4.20609712600708, "identity": 0.24415382742881775, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0d569a49ddb3363de89880902538b947a4aa3d14163b85236e9d20cb2e5fb4f4.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "0d672796a1d6e592405e571cd381817f2fee3ee171669609b515dbdaaae7d81d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "807489611823df8d5a58e15e57e98e9a3c89f50d88821ba2f2087e411e2dc072", "generation_seconds_batch": 22.671761453035288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 5.998785495758057, "genuineness": 3.835810661315918, "identity": 0.6588429808616638, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0d672796a1d6e592405e571cd381817f2fee3ee171669609b515dbdaaae7d81d.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "0da2ed795a91e9196f9203dda4aa45d75e99f211a3ab19a8c5355550ba93e018", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ce18b8776e52434fae5262103fd5225cdf25faf6aea13d6f9ad2050a1bc95187", "generation_seconds_batch": 22.145384954987094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.09409761428833, "genuineness": 4.118232727050781, "identity": 0.6879199147224426, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0da2ed795a91e9196f9203dda4aa45d75e99f211a3ab19a8c5355550ba93e018.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "368bf9eb18d5656b8dae0bd35ab9c5a77f0ec44ecb666a9a7068acfabd223341", "generation_seconds_batch": 17.764571076026186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.1020402908325195, "genuineness": 4.6203155517578125, "identity": 0.727643609046936, "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_WARM_reg_high/0dba5cebf35b1ddca791f8d077041a7c94b74d09f6947046262df9fe6084771a.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "0ddd099dd90880c3eacd645c5593a04ac5cee635b5e81618254437024dcf95fc", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "229e6801fc8ea77c834cd1e5c1451e677f7d281f7906457f6801261869981238", "generation_seconds_batch": 16.483222765033133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.181706428527832, "target_z": 1.7421873106178378, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.7214093208312988, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0ddd099dd90880c3eacd645c5593a04ac5cee635b5e81618254437024dcf95fc.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "0e7b219dffbbcfee353ae0ddc6348d82c2f458a27327bae75c443adc3e84e741", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2a5c3e59ddf4bc53e09b28701001715aab4a55b36ced215fa42291a50aa5a5ff", "generation_seconds_batch": 25.92831645498518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.205295085906982, "genuineness": 4.099404335021973, "identity": 0.7283757925033569, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0e7b219dffbbcfee353ae0ddc6348d82c2f458a27327bae75c443adc3e84e741.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "0ee18b6a92e0880dfb63e52767b96a29ceeb66cfb12cd4bfd1c6633a5df2293e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "166c70980d36208641de93c776b625fcb2360d6ef7b1653e8075df1565fadae3", "generation_seconds_batch": 21.136285643093288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.404097080230713, "target_z": 0.8828125576380493, "ce": 6.260893821716309, "genuineness": 4.086852550506592, "identity": 0.18060705065727234, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0ee18b6a92e0880dfb63e52767b96a29ceeb66cfb12cd4bfd1c6633a5df2293e.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "0f41e8daf594fdfa8ae658626ce7dc0f71bd591e8dce7c7ebfb81dbb9074c417", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "577d3105333f48f2e14ea5291c5e8c22f20e974c31297d80f669a1def571d06b", "generation_seconds_batch": 16.961068186094053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4394428730010986, "target_z": 0.9218749505959577, "ce": 6.284721851348877, "genuineness": 4.406930446624756, "identity": 0.4180551767349243, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0f41e8daf594fdfa8ae658626ce7dc0f71bd591e8dce7c7ebfb81dbb9074c417.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "0f46f68290cc8e07d9f5fb994358c2a026b3bcfd382c124adf53d56c168b84d8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "cef9d91acccff5ed7e0aa2567d32501a83c311d216bca0e7ffe39a75ab777428", "generation_seconds_batch": 24.42609712597914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 5.303802013397217, "genuineness": 4.432034492492676, "identity": 0.6708939075469971, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0f46f68290cc8e07d9f5fb994358c2a026b3bcfd382c124adf53d56c168b84d8.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "0f4e163edbd4012c01dafa127a7135e4afc84415019d6bd35ab01be076f14315", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "9f21f295487abec7be4f8dff6cc22809fe2110f7b4e884ebad69c0c2eb08ca3a", "generation_seconds_batch": 24.714191720006056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.213237762451172, "genuineness": 4.105680465698242, "identity": 0.2921985387802124, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0f4e163edbd4012c01dafa127a7135e4afc84415019d6bd35ab01be076f14315.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "0f54188853c70debb2ea4e2625d1e330984c63778f1191b923632a62f2486d6f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c02b3bcd5bcb978ee6a13ba7034f7b5bbbee5a48b108dd73c57ebfe62a2c9cde", "generation_seconds_batch": 23.39201901596971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.38288950920105, "target_z": 0.8593750164680141, "ce": 6.117925643920898, "genuineness": 3.860914945602417, "identity": 0.37679144740104675, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/0f54188853c70debb2ea4e2625d1e330984c63778f1191b923632a62f2486d6f.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "0fa863e9bfb64f1de4c10b244efbdef14859189b266293427ba08f8e5bd43c02", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "83e7f78ebc3fd16206a4b76d8f31a66394e21c7297bbd287d8b370da12c3ef23", "generation_seconds_batch": 23.88352804200258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.133810997009277, "genuineness": 4.7458367347717285, "identity": 0.6860544681549072, "hypothesis": "J'ai trouvé la 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_WARM_reg_high/0fa863e9bfb64f1de4c10b244efbdef14859189b266293427ba08f8e5bd43c02.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "108f90214ef66a05bee296b9c96d35354a40a7dac75c50c8e8d87735a1e6945b", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b1d2b14ac093b1fc86430600a6cff4f62b32d8903be4dab87fc8fe2c4d162bd3", "generation_seconds_batch": 17.237253520986997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.305128574371338, "target_z": 0.7734375411700353, "ce": 6.173524379730225, "genuineness": 3.9550557136535645, "identity": 0.2947235405445099, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/108f90214ef66a05bee296b9c96d35354a40a7dac75c50c8e8d87735a1e6945b.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "10a071a4a258a6c975e802850b3276e48b16e8cfa0371ae3a6586d00d93dfa75", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "aa12e195776abe0336d5b54b6c489c464df822072fc1bd662fb566ef665648fd", "generation_seconds_batch": 16.47887192806229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6727256774902344, "target_z": 1.179687376489894, "ce": 6.221180438995361, "genuineness": 4.293961524963379, "identity": 0.6998425722122192, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/10a071a4a258a6c975e802850b3276e48b16e8cfa0371ae3a6586d00d93dfa75.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "10ad6f333329f93401b0a7ce697b471a670dab5d26897a6cdd61ae32e0a51bcf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2d65fb0866865d1b12a0f46f24b677d7e416bc64710e9c8dd74970a942bb1412", "generation_seconds_batch": 24.779135588090867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.419747352600098, "genuineness": 4.293961524963379, "identity": 0.6207377314567566, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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_WARM_reg_high/10ad6f333329f93401b0a7ce697b471a670dab5d26897a6cdd61ae32e0a51bcf.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "115a55533eb8e07d2236ce5ae395ba43d50bb936db3499e118863c72b7678f33", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d009407cdee531b0dcc756936d8cb69a69eafdf802b9f0ec0bfe7d9f80b4971d", "generation_seconds_batch": 19.014681337983347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3546128273010254, "target_z": 0.8281250494040423, "ce": 6.062326908111572, "genuineness": 4.287685871124268, "identity": 0.6090825200080872, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/115a55533eb8e07d2236ce5ae395ba43d50bb936db3499e118863c72b7678f33.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "12674699f07bd1763ba3a2cf21e583e4ca6e40f3e59e21e5ea609da743f61bea", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b3df245f0bca915c22be05d0e5343060aeb6f2259cf4e6064adcb13ca288725", "generation_seconds_batch": 20.43076894094702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.268836498260498, "genuineness": 4.557555198669434, "identity": 0.3779999017715454, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/12674699f07bd1763ba3a2cf21e583e4ca6e40f3e59e21e5ea609da743f61bea.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very warm vocal delivery. 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": "f579afe2839b6a5ad0f14376191c07b175a4017a98d333b9f6bfab3d317055a1", "generation_seconds_batch": 19.497521262965165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.157639026641846, "genuineness": 4.457139015197754, "identity": 0.6495933532714844, "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_WARM_reg_high/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "13121485c0914d6d6b37a752c49319b3f3790c54d839c92aed0a6f8bc27e9c0d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "100ae638c114a1e6c86b5af9df59a92bc8cb1dce2a278770a43b1572a8126d6a", "generation_seconds_batch": 21.05181061697658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729279041290283, "target_z": 1.2421873106178378, "ce": 5.919358730316162, "genuineness": 4.055472373962402, "identity": 0.563347578048706, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/13121485c0914d6d6b37a752c49319b3f3790c54d839c92aed0a6f8bc27e9c0d.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "134c458ecc0bce64e0a5c3f1b5eba88a306561fbbd58876ce0548ca2a0ca2a79", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "157d6afc735bcd12bd71d38fc6f4123ba7b07c71fdbfe931ade126e849e7499e", "generation_seconds_batch": 23.451818525907584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9554929733276367, "target_z": 1.4921875741060635, "ce": 6.300607204437256, "genuineness": 4.356721878051758, "identity": 0.39684516191482544, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/134c458ecc0bce64e0a5c3f1b5eba88a306561fbbd58876ce0548ca2a0ca2a79.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "13685e8deef9fe3840750eb5f65ddf9f7b3b16650aa7c006215f5ebf454001be", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "eb0c73df55cace4762a6ea9854232d7467c031b19b830c3c6f087a5551d19024", "generation_seconds_batch": 22.782345279934816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.356205940246582, "genuineness": 4.143336772918701, "identity": 0.7261074781417847, "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_WARM_reg_high/13685e8deef9fe3840750eb5f65ddf9f7b3b16650aa7c006215f5ebf454001be.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "13e6d8344021ca418b1712918347d9f2f61f0a486d53f620fd7b5485b71c1a4c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "611f08edc1ed8321f2f17853a2ef5ba94b33b564347d8621f82eba32fcbbf697", "generation_seconds_batch": 24.00880176492501, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.7442495226860046, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/13e6d8344021ca418b1712918347d9f2f61f0a486d53f620fd7b5485b71c1a4c.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "142218099487d5b96cae0eb16b73f44e2f24cf4439a24ea4f406a630ba3244d1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8704dd1c95daa85d7a2c6d4151ce62132f70446ddbacbabed81e916d37d23996", "generation_seconds_batch": 17.78092242591083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.070269584655762, "genuineness": 4.030367851257324, "identity": 0.5626024603843689, "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_WARM_reg_high/142218099487d5b96cae0eb16b73f44e2f24cf4439a24ea4f406a630ba3244d1.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "142cb9c1566429b7c96bb0b70990e9faa7027d8464843dcbb2a8c18919735572", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bf60c324c9ef55ba2ddbda815878f3e525725d0b1cb54dd14e4de5b59eee336d", "generation_seconds_batch": 23.521052537020296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729279041290283, "target_z": 1.2421873106178378, "ce": 5.8875885009765625, "genuineness": 4.507347106933594, "identity": 0.6787652373313904, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/142cb9c1566429b7c96bb0b70990e9faa7027d8464843dcbb2a8c18919735572.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "1441648e8064ddf59022dcb3750e15fce1405d9a1a33f54c709be73b48afb0c0", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2e6e4abcb8791b3fd0c29c5ee881217a3a0824d729ade465d4af5b1137f01c22", "generation_seconds_batch": 17.930977203068323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.7818554639816284, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1441648e8064ddf59022dcb3750e15fce1405d9a1a33f54c709be73b48afb0c0.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "1455c53047ed16b867ca54e308f172cddc7322a72fe53ff67c3170eedc756b07", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1ea5a70466ce13f773433df6f1de0b2483e1004f92d8a75394ad95f30c681453", "generation_seconds_batch": 14.402103102067485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.213237762451172, "genuineness": 4.394378662109375, "identity": 0.7189012765884399, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1455c53047ed16b867ca54e308f172cddc7322a72fe53ff67c3170eedc756b07.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "90c3e6a98c8c0f3b2c80215df649545a5a5ae029a2c9c7fcf8a4bd14760a7bf6", "generation_seconds_batch": 21.184748630039394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.7105427384376526, "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_WARM_reg_high/14849f3a945a1e8909ad42ab4be19af91b7a426d0bb5f8a29c0f85ca6a49e479.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "1498f04698b8f43fa4bc46ec2a53a0d43cee67a6b192f67dde8f89ea229cb993", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "690e0e9c04df8fbd7dc343206ae1e7e8949fecde70090c08e8be3884c23310b7", "generation_seconds_batch": 16.346237158984877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.41180419921875, "genuineness": 4.017816066741943, "identity": 0.761107325553894, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1498f04698b8f43fa4bc46ec2a53a0d43cee67a6b192f67dde8f89ea229cb993.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "14b9ee71f70965ce42d69b460e8d768e997ceb2c193e52619e7d6272d43a8949", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "94965603af75ec64fccb71f4f473853348c3859e8eb4be6d465e3fc529b39199", "generation_seconds_batch": 16.321040828013793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.312197685241699, "target_z": 0.7812499670639718, "ce": 6.252951145172119, "genuineness": 4.607763767242432, "identity": 0.7745736837387085, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/14b9ee71f70965ce42d69b460e8d768e997ceb2c193e52619e7d6272d43a8949.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "14c2324ab215a45ee6b46952fac479712c122ab663e9a3226e1add642878b820", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5d77e9d2472f5edef8952ba80ecfed2bf07ce68be0052b8c7625e9c3a7edd44c", "generation_seconds_batch": 25.27418491896242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.026184558868408, "target_z": 1.57031236002188, "ce": 6.252951145172119, "genuineness": 4.256305694580078, "identity": 0.37321147322654724, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/14c2324ab215a45ee6b46952fac479712c122ab663e9a3226e1add642878b820.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "06516a0ae0da3b590ec40ae6d9890b70b567c93e3daed05eeb0eda554f5265e8", "generation_seconds_batch": 18.26155725494027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0120463371276855, "target_z": 1.554687508234007, "ce": 6.443574905395508, "genuineness": 4.381826400756836, "identity": 0.5815410017967224, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very warm vocal delivery. 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": "e2a8b464ced97942efe9c3c57336c5ed79ece4ae842c3870d71cf8fda60a68bb", "generation_seconds_batch": 18.985152882058173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.300607204437256, "genuineness": 4.168440818786621, "identity": 0.3583318591117859, "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_WARM_reg_high/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "1602f181f19ca831e96bc44a30124cb2d45ac87337580e03d7e3477dd7ccba9d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "7fd9ec086f6642b6600fefd347561952ae704f9d4a4769d2fbf2790093a8125e", "generation_seconds_batch": 22.412906074081548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 5.903473377227783, "genuineness": 4.632867813110352, "identity": 0.7871646881103516, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1602f181f19ca831e96bc44a30124cb2d45ac87337580e03d7e3477dd7ccba9d.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "1614240ea7508d5484077673bfeb319e1a6a8d46fa0a20237da1daba3c7da5c6", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ecbad344b48b3195a2fdc3dc68f47c44517e55b25a4cef52ff53ff5b19c9c72e", "generation_seconds_batch": 16.346237158984877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.268836498260498, "genuineness": 4.356721878051758, "identity": 0.6355684995651245, "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_WARM_reg_high/1614240ea7508d5484077673bfeb319e1a6a8d46fa0a20237da1daba3c7da5c6.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "165821280b3df97324cc1e45b199501ad3ff6e77fb9600801112f56f49ff2270", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "18c9b9ce6bc26b6e674968a8a39617878aa216cab9bb9738c8b815eec8be7ce6", "generation_seconds_batch": 17.57428386399988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.221180438995361, "genuineness": 4.469690799713135, "identity": 0.6785911321640015, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/165821280b3df97324cc1e45b199501ad3ff6e77fb9600801112f56f49ff2270.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "16727b25fd3c3cd0ccb118f9754b1cb39c37d2123ff75ba81a8d2cf9747670bc", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "07b17b6c625cde8b4ffcbafd6753675fe27f3e16d85e491be58ce06eeed36823", "generation_seconds_batch": 14.758363711996935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.1894097328186035, "genuineness": 4.281409740447998, "identity": 0.619025468826294, "hypothesis": "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_WARM_reg_high/16727b25fd3c3cd0ccb118f9754b1cb39c37d2123ff75ba81a8d2cf9747670bc.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "17a4bb39b5f8c8b5165dea0533817055a4b45df862e4a18424136d8c3574eb14", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fffb194a0811aa1fe2754869636d0ac6d339a024a481a383778f6f276ad94e36", "generation_seconds_batch": 22.840937337023206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.356205940246582, "genuineness": 3.9676074981689453, "identity": 0.5342401266098022, "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_WARM_reg_high/17a4bb39b5f8c8b5165dea0533817055a4b45df862e4a18424136d8c3574eb14.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f3ad6c868f4c0728ef99fc4c2109617e103f107f862a9af5e859cc62caaefaae", "generation_seconds_batch": 21.213882546988316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.046441555023193, "genuineness": 4.494794845581055, "identity": 0.620750904083252, "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_WARM_reg_high/17d6cfed5f2430a0c2bb15ec39aa5af7e62f3a00c0e9cf9f29eb86219e136166.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "10c6cd47691d836c3b7d5b7b8829b32d758ed8f069d6854349b5fbf13e5dfd23", "generation_seconds_batch": 23.527283388073556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.0146708488464355, "genuineness": 4.494794845581055, "identity": 0.7387996912002563, "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": 0.120000000000001, "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_WARM_reg_high/18647642260d038b37f2c356f502f1973d7db2bc9128aeb5ba3edb0c169a50c5.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "18a8a38d64f833465107d343dc989ea4984dc3b19fa529b407850487b26f1899", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b246e27cf71cb8212714bf3d256df0c9adb86136adb286eb073ab5f089ead28", "generation_seconds_batch": 24.327557830954902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.229123115539551, "genuineness": 4.1998209953308105, "identity": 0.1606433391571045, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/18a8a38d64f833465107d343dc989ea4984dc3b19fa529b407850487b26f1899.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "18c304cd21704cb064d0215f62e7ca4eff68cc8d65faff757ebb89621d71753c", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2a58d812accd2433d81c32ab47370d5c38b9dedc345145da4d51e0312350230d", "generation_seconds_batch": 17.64097264909651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.414700746536255, "target_z": 0.8945311964789542, "ce": 6.046441555023193, "genuineness": 4.432034492492676, "identity": 0.8142301440238953, "hypothesis": "La porte 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_WARM_reg_high/18c304cd21704cb064d0215f62e7ca4eff68cc8d65faff757ebb89621d71753c.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "18e46c662f0d7384566f80785a4cdce5649e02cd7ac044e2e477d222905d6913", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9b1a09f0a9c69ccc403cf7f3edfc31dee8e91228e888634e232ea405936938bd", "generation_seconds_batch": 24.774293038994074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4606504440307617, "target_z": 0.945312491765993, "ce": 6.427689552307129, "genuineness": 4.582659721374512, "identity": 0.5358414053916931, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/18e46c662f0d7384566f80785a4cdce5649e02cd7ac044e2e477d222905d6913.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "195020bf9b6fceadedbbf3179300027cc34c237ac5585caf81ce772a6cf7a139", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e6922b5c620fd100026018db4388be2dea64851a8e97f7f9bf6ba524c7b33b86", "generation_seconds_batch": 23.54294746601954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.300607204437256, "genuineness": 4.15588903427124, "identity": 0.6979081630706787, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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_WARM_reg_high/195020bf9b6fceadedbbf3179300027cc34c237ac5585caf81ce772a6cf7a139.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "1ae1df8f4aaba73983426bd1bd8680bbb1cf88d90eecef2ac0ea5cdacb924197", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e0fb936d6bcd0a055ba3c070bf5d51bbcc29b53da75cf2a1ea9d7745eca91820", "generation_seconds_batch": 20.922719527967274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.227367639541626, "target_z": 0.6875000658720565, "ce": 6.117925643920898, "genuineness": 3.7793264389038086, "identity": 0.30919361114501953, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1ae1df8f4aaba73983426bd1bd8680bbb1cf88d90eecef2ac0ea5cdacb924197.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "1bb75b033a61d367bc37be7ac9d48b32d29b536603bb03257dc5e5df9ad458e3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "676113711469925a59626aab5cde6d43874346c426c1753458ef5febe04447a7", "generation_seconds_batch": 17.23486279696226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.09409761428833, "genuineness": 4.413206577301025, "identity": 0.6196615695953369, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd: zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1bb75b033a61d367bc37be7ac9d48b32d29b536603bb03257dc5e5df9ad458e3.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "1be5f8197a32a133b44d6fd6eabea606855b0f21eeade8740745aba0e1d5e032", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "80a960a693d6870d479477787bc7f2e18a762bb81ed05ef97cf9ee9198335073", "generation_seconds_batch": 20.922719527967274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.348263263702393, "genuineness": 4.20609712600708, "identity": 0.47953343391418457, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1be5f8197a32a133b44d6fd6eabea606855b0f21eeade8740745aba0e1d5e032.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9de13747e751f14da3dba27ab41579614f7e72138135c450ed868ede68636806", "generation_seconds_batch": 18.942736728000455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.096876621246338, "target_z": 1.6484376729141481, "ce": 5.95112943649292, "genuineness": 4.582659721374512, "identity": 0.7940037250518799, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis les biens et vérifie les deux derniers chiffres. Appelle-moi quand y aura un moment.", "wer": 0.18181818181818182, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1ca508c27b108a98ee165cfb7dab64134f3a219b8f8ab7bd25d969b67ebb7cad.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "1d5358954bb2a1a0cbc3b5b20cc9472eebb3ea07794f7c78f5e169cfdba0883f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "37edc14569f63e2fc59f0c8cb84cd77f3c9ac9e3266a1352c237eb27316482e0", "generation_seconds_batch": 23.777553699095733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344297409057617, "target_z": 1.921874687107732, "ce": 5.990842819213867, "genuineness": 4.124508857727051, "identity": 0.6314929127693176, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1d5358954bb2a1a0cbc3b5b20cc9472eebb3ea07794f7c78f5e169cfdba0883f.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very warm vocal delivery. 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": "706846a33a64bbe4ef73be7249a558e8e0e41bbf1f317a3870be9b34cb96e494", "generation_seconds_batch": 16.699607365997508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.6571103930473328, "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_WARM_reg_high/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "1d6c358c5853ffa7d9ab6d314e2090333e91f285389cf84e02f91aa26a5b448d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4daf97f774a1b41a0e21875dae18e0a5e490b0c341b98193ba93e5d5aa51d156", "generation_seconds_batch": 17.386325592990033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.308549880981445, "genuineness": 4.093128204345703, "identity": 0.5165505409240723, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1d6c358c5853ffa7d9ab6d314e2090333e91f285389cf84e02f91aa26a5b448d.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very warm vocal delivery. 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": "4a1b5a0edb9c45e05bad239af895b7dfff7c3903780f734fffd98b94660ac6e3", "generation_seconds_batch": 18.52323825797066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.348263263702393, "genuineness": 4.545003414154053, "identity": 0.6124590039253235, "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_WARM_reg_high/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7cc8eb8166c90b0c988bcf71bc0b0f51d3062b2f5005134ca7415e6c7e875cd7", "generation_seconds_batch": 24.33590683399234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 5.974957466125488, "genuineness": 4.369274139404297, "identity": 0.6798831224441528, "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_WARM_reg_high/1dbbd10fc82e0ac79c76be68ce19cfd3206cf493528dba4c456b46cb9b9971fb.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "1e06cf0431e188ff5fb3878ef66d870fa9f01a603a2dcece3346193be9825dae", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3be32cca8070f522506ef3d67d85a7e5f0208787907c46ad539d312c3572c7e3", "generation_seconds_batch": 24.464731225045398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.395919322967529, "genuineness": 4.281409740447998, "identity": 0.3158129155635834, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1e06cf0431e188ff5fb3878ef66d870fa9f01a603a2dcece3346193be9825dae.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "1e417a1ae3e6cf349900888880ae924808e20817d6459afacf061cf4b42de212", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "86402f83849b9ccd90df0e3785ae8cca6f78f8d37d210af0fbf063bdc5598334", "generation_seconds_batch": 23.38369412207976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.443574905395508, "genuineness": 4.419482707977295, "identity": 0.2776787579059601, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1e417a1ae3e6cf349900888880ae924808e20817d6459afacf061cf4b42de212.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "1e68951795e1465e010d836bc80a8926ae6694e9788e16f2a476d39b3d830f44", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "eddeaaa1f32ab5240b20b2cc3346a0e107c7f09b5cb902873805be0b32ea1b46", "generation_seconds_batch": 22.86091861105524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 5.998785495758057, "genuineness": 4.18726921081543, "identity": 0.6574659943580627, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1e68951795e1465e010d836bc80a8926ae6694e9788e16f2a476d39b3d830f44.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "1e88d00b2daa3423c91ec72d206a29221c626b579fb2ad187eb1a2bb9b387e1d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f282a7f45c1649ed3eec4b81a524f04a6e4a296dee5e33d75a77059789aa0627", "generation_seconds_batch": 23.73602718207985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.41180419921875, "genuineness": 4.080576419830322, "identity": 0.5796608924865723, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem altern Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1e88d00b2daa3423c91ec72d206a29221c626b579fb2ad187eb1a2bb9b387e1d.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(very warm vocal delivery. 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": "afae0bc9df98909a43525a5aed1a98696f9b680c7438cc7d33123828caec1266", "generation_seconds_batch": 16.699607365997508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.268836498260498, "genuineness": 4.105680465698242, "identity": 0.6887996196746826, "hypothesis": "Describí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "1f39a7ef5fc1ca3adf6a7e2493b71c989afbeaf91e85ceeaf63e90a033d14a00", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "00cbf9ffa0665bd12c1499193c9e73ca2e1646a941985bca84f87120995447ed", "generation_seconds_batch": 23.17728814308066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898939609527588, "target_z": 1.42968763997812, "ce": 6.538887023925781, "genuineness": 4.3881025314331055, "identity": 0.6164788007736206, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1f39a7ef5fc1ca3adf6a7e2493b71c989afbeaf91e85ceeaf63e90a033d14a00.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "1f4711236428767921307a319217480ccbc67fb9f859045a4b046a6e30e4545c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d5c058353da6e5c06f5df355d1e618e96ec24f3ba8e6d8cc52be04b01df9751b", "generation_seconds_batch": 25.501212641014718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.1894097328186035, "genuineness": 4.306513786315918, "identity": 0.7353324890136719, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1f4711236428767921307a319217480ccbc67fb9f859045a4b046a6e30e4545c.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "1f97938f637c5e72c44984092ce9e7586de5c524d768c3f51fa607745f5c9c27", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d9bdccb4d32ea00c9f26031b291ef883013cf0991273e327954d082c4e7064c1", "generation_seconds_batch": 17.66231753793545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686864137649536, "target_z": 1.195312491765993, "ce": 6.4515180587768555, "genuineness": 4.482243061065674, "identity": 0.2855333685874939, "hypothesis": "I found a little box. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1f97938f637c5e72c44984092ce9e7586de5c524d768c3f51fa607745f5c9c27.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "1f9ae5cfa406b451395fd5f85afa30c878737c6cc208468e94d3d69a80bb0e87", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b00eaa0a068529962eeb8ae0b3b8bfa61d8d85bc8d3567a87776e4c4afb541a6", "generation_seconds_batch": 17.16836826596409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446512222290039, "target_z": 0.92968763997812, "ce": 6.1894097328186035, "genuineness": 4.086852550506592, "identity": 0.1798730492591858, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1f9ae5cfa406b451395fd5f85afa30c878737c6cc208468e94d3d69a80bb0e87.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ae8213367f1dd5579eb448dbcb490ff3156e33dc0993533eca139ea5568911e6", "generation_seconds_batch": 24.08911552594509, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.213237762451172, "genuineness": 4.243753433227539, "identity": 0.6962215900421143, "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_WARM_reg_high/1fa4ae67436ee7a5db578e016c85856564edf45fc1d5c89fbddfcb213ca9cac6.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "1fdc302a0f353bdc3cc34c16118ac58e199f16a18efb262ed5555d1a6ae938d9", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "de01e6653af15ddd1680dabe815b6cccec072f2930bbe8d5cc47e6ce58825e15", "generation_seconds_batch": 17.35467329097446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.1894097328186035, "genuineness": 4.306513786315918, "identity": 0.559429407119751, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/1fdc302a0f353bdc3cc34c16118ac58e199f16a18efb262ed5555d1a6ae938d9.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "220b81bbc6a95d12697285be6571ba5ad91c52220c73d49db1105b106de12da1", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8a4dd1180270e6c6cfbe8ec7d44b8d9c93ad149ba7643fb735d0531b2476f187", "generation_seconds_batch": 15.946987231029198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.3641486167907715, "genuineness": 4.111956596374512, "identity": 0.3882554769515991, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/220b81bbc6a95d12697285be6571ba5ad91c52220c73d49db1105b106de12da1.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "81adb0de2e3ef2e01fa9449e11d4ad30a9a208c434991bd1324d4a423a0144ae", "generation_seconds_batch": 25.65294576191809, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.828247547149658, "target_z": 1.3515623270858519, "ce": 6.308549880981445, "genuineness": 4.093128204345703, "identity": 0.4223138689994812, "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_WARM_reg_high/2237109725e3ec0ba88f5346e691f08c3878b8e333a85a3e51903ef8c40acdb1.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "223bcbeccc40f2c38ec5a3323cbeea3e9844fcf254e700035c7502a668460e69", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "2feae293d34a2c794ac1d6d8a604d920d77e15b1b1854ce1c395ba96309ef442", "generation_seconds_batch": 23.86341965198517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.181706428527832, "target_z": 1.7421873106178378, "ce": 6.205295085906982, "genuineness": 4.720732688903809, "identity": 0.5550815463066101, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/223bcbeccc40f2c38ec5a3323cbeea3e9844fcf254e700035c7502a668460e69.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "2259c66c9520dd780334177aa29d39d6af9ee993a17fea151d6f2e1be06f1b02", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "66dea8b59a859769a1bb666e08de808c017b1fb43ffb752e2be66ef3c9b6e5d9", "generation_seconds_batch": 23.521052537020296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4076316356658936, "target_z": 0.8867187705850177, "ce": 6.419747352600098, "genuineness": 3.948779582977295, "identity": 0.5391218662261963, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2259c66c9520dd780334177aa29d39d6af9ee993a17fea151d6f2e1be06f1b02.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "2351832e6b48f723eac093db093665eab56b9db0dbd01c2a54517b56e7b369dd", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9e6b35389b8aca576519ae2f876648be4bae34faa7b17cddcc7c40705c91f36c", "generation_seconds_batch": 16.09815354901366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.0146708488464355, "genuineness": 4.174716949462891, "identity": 0.4433441758155823, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2351832e6b48f723eac093db093665eab56b9db0dbd01c2a54517b56e7b369dd.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "23c1150b879ff682e622c38b9da93994880badba36b720a85de23e644d0878f1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "110a5a512859109a29f0aac390d6ae5d3139bc576e2315c64367c8e19a15639e", "generation_seconds_batch": 23.44479463400785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.38003396987915, "genuineness": 4.337893962860107, "identity": 0.3564561903476715, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/23c1150b879ff682e622c38b9da93994880badba36b720a85de23e644d0878f1.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "23cddfb98a7d9814fbd7c7e11bae8856d5d7224d4fc99153b4666fc0de23c7aa", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ae03e2dfbd57ee6561b29ece4e9905401f259f6a8177ab9d3e45b7b03df729c8", "generation_seconds_batch": 23.97997178300284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.24500846862793, "genuineness": 4.432034492492676, "identity": 0.7097885608673096, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/23cddfb98a7d9814fbd7c7e11bae8856d5d7224d4fc99153b4666fc0de23c7aa.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "23ddc92a0a1de361c167dd5d4c60305f21fca93c75253a787b231a1a48e84d43", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fefa857f5e444b23840add3e4ffe206649dd82d39ba4519f5470e14e307c3cc2", "generation_seconds_batch": 24.670562471961603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.125868320465088, "genuineness": 4.293961524963379, "identity": 0.7206103801727295, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/23ddc92a0a1de361c167dd5d4c60305f21fca93c75253a787b231a1a48e84d43.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "23f084d643c0f34f6f2f651b136496aa78579872e460c59b8c48a20464c38af8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "053e73b92624ade5a4b00ac531d99df21fc2dccbabebfa7f1a5ce50a68013cd8", "generation_seconds_batch": 24.327557830954902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.403861999511719, "genuineness": 4.394378662109375, "identity": 0.21256974339485168, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/23f084d643c0f34f6f2f651b136496aa78579872e460c59b8c48a20464c38af8.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "2423b8e9f7864fe66e49b5d9500fea9e0d545c161857b95c5b92dc08308040b6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1db1545f6a8d4d3d8a23b7e84a08b45a8e03d3bfb2e5b7d28a6daf4297f0abd6", "generation_seconds_batch": 22.468655899981968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 5.538110733032227, "genuineness": 4.469690799713135, "identity": 0.7581806182861328, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2423b8e9f7864fe66e49b5d9500fea9e0d545c161857b95c5b92dc08308040b6.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "2474930d84f62b025d0d88acac0a4c2b34510d53d744c3b053bf90584f258f90", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fa8103e17c661f508701e03455a6ee2ffeef8385384f9691bccddb61181dbf99", "generation_seconds_batch": 16.255138618987985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4747886657714844, "target_z": 0.960937343553866, "ce": 6.2767791748046875, "genuineness": 3.7918784618377686, "identity": 0.36319443583488464, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2474930d84f62b025d0d88acac0a4c2b34510d53d744c3b053bf90584f258f90.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "25146af878e3d9e0b02dbb3627a0f4e6e393e2db03f012f77c46158835e25748", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "55ebe61df6826ef45d6c4ea6085c0e9c830dbe37d9ed28792ff71456214fc4c8", "generation_seconds_batch": 22.521420850069262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.884801149368286, "target_z": 1.414062524702021, "ce": 6.308549880981445, "genuineness": 4.300237655639648, "identity": 0.324049174785614, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/25146af878e3d9e0b02dbb3627a0f4e6e393e2db03f012f77c46158835e25748.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "253aacf80660b5a054441baee98e78c1ed513a7d4b2d863c9455fcb98a3d0a96", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fb015c1f0f16237b6f430ef27362db96ef7c8e07bdf758c20d24cfb862a69e07", "generation_seconds_batch": 17.469084804994054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.118083953857422, "target_z": 1.6718749505959576, "ce": 6.308549880981445, "genuineness": 4.532451152801514, "identity": 0.601820707321167, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/253aacf80660b5a054441baee98e78c1ed513a7d4b2d863c9455fcb98a3d0a96.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "25b62ef1fd793b302d20547969bde9dde2f29066b42ceeaf28edfc357fb16503", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fee515b9055b00adfa14b22b9d60fb0f33c22d68dac313cf840a2d9e7f002680", "generation_seconds_batch": 17.21395555092022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.062326908111572, "genuineness": 4.381826400756836, "identity": 0.6935964226722717, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/25b62ef1fd793b302d20547969bde9dde2f29066b42ceeaf28edfc357fb16503.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "25c1c27e2a0cbcfa45f56dce6c2dbffd2a5b7cabe3fc6413c317c2e58e901ca9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a02d037678ad42443e8d6ee3f709284930feb5f5857a9eb54725ec2f9725d216", "generation_seconds_batch": 23.03621890710201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4747886657714844, "target_z": 0.960937343553866, "ce": 6.197352409362793, "genuineness": 4.068024158477783, "identity": 0.7323624491691589, "hypothesis": "De deur 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_WARM_reg_high/25c1c27e2a0cbcfa45f56dce6c2dbffd2a5b7cabe3fc6413c317c2e58e901ca9.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "25fd174342ff39000ed77889786b9070ca21fff56dfd4b99871363b4e89232f5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "85f6784a2b5b035cb00f1b65554c671ee9e721305da16a362be52b6fdebc0bb5", "generation_seconds_batch": 17.723066051956266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.488927125930786, "target_z": 0.9765624588299647, "ce": 5.831989288330078, "genuineness": 3.7416701316833496, "identity": 0.6003561019897461, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/25fd174342ff39000ed77889786b9070ca21fff56dfd4b99871363b4e89232f5.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "261975b9ec8c62be1907d3f258392a6a6c40ef623fba75a8baf2f48d507433b1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "893182ce291162e57bf9a26ecdfe1a54632a25425ff7a550f904ddb7809d25db", "generation_seconds_batch": 24.413496519089676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8565244674682617, "target_z": 1.3828125576380494, "ce": 6.173524379730225, "genuineness": 4.306513786315918, "identity": 0.684554398059845, "hypothesis": "De deur 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_WARM_reg_high/261975b9ec8c62be1907d3f258392a6a6c40ef623fba75a8baf2f48d507433b1.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "2644601dd358ff811f5647703f742729beb4338676408ade3832fbc4a3715292", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "12af3e86984d2bd6f562fa6d449e844def1c281f5f6f7ce602c244e5254cd545", "generation_seconds_batch": 22.68763935903553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.372574329376221, "target_z": 1.9531249176599295, "ce": 6.022613525390625, "genuineness": 4.519899368286133, "identity": 0.5841430425643921, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2644601dd358ff811f5647703f742729beb4338676408ade3832fbc4a3715292.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "264e8765d8393fead39d517306963b88622fe29db59d487a1b935e1dcc2f03d8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4b454184bc05245eb958999a8ee21eb79aefca1f14ec8eac8599d779fcca56c3", "generation_seconds_batch": 22.080510414903983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.221180438995361, "genuineness": 4.457139015197754, "identity": 0.6371179223060608, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/264e8765d8393fead39d517306963b88622fe29db59d487a1b935e1dcc2f03d8.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "265c044224a2ac8a2a3d8002f463a06561a0c27fda3b7bc829903a7fc90b32d3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f3e7105686fc8cf4dc1d224d6a1b2caa2720e15cbc17d17efb339637254ee3b3", "generation_seconds_batch": 17.562399860005826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411166191101074, "target_z": 0.8906249835319859, "ce": 6.38003396987915, "genuineness": 4.413206577301025, "identity": 0.6761707067489624, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/265c044224a2ac8a2a3d8002f463a06561a0c27fda3b7bc829903a7fc90b32d3.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "26922ef9fc97fa2da285a623f98949ef8d389ce52f5f614778a672ca5f7415e3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "19d34d78041da91c6153bed97c60e710624d278228cffaee359ca1f000d46064", "generation_seconds_batch": 27.359727213042788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.217052459716797, "target_z": 1.7812499670639719, "ce": 5.935244083404541, "genuineness": 4.394378662109375, "identity": 0.5838459730148315, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay detrás.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/26922ef9fc97fa2da285a623f98949ef8d389ce52f5f614778a672ca5f7415e3.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very warm vocal delivery. 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": "391bd9921384e5d62f0ab6632f8656d93ca121ed4e149ed95833e6ddadb60201", "generation_seconds_batch": 17.584979100036435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.38003396987915, "genuineness": 4.036643981933594, "identity": 0.2938016355037689, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "adeb123f46fd13142c44c537dbacc52e7074248d063a8c82ea0358881a8c0b1b", "generation_seconds_batch": 24.210018084966578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.118083953857422, "target_z": 1.6718749505959576, "ce": 6.221180438995361, "genuineness": 4.3504462242126465, "identity": 0.7750614881515503, "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_WARM_reg_high/26d586e1e4c326b0203dc97d608ee1e7fa64d6ddd2402c370534768d153c3016.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "26dc27efb52bc7c469e00e018380776d8b69a41b9e21a6215c51bdd6b30b1fec", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d9ed52dc68d398da98a20304824f04e06d2d58ae1c193876a52203df34e4d376", "generation_seconds_batch": 15.87749464600347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.109982967376709, "genuineness": 4.174716949462891, "identity": 0.6733075976371765, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/26dc27efb52bc7c469e00e018380776d8b69a41b9e21a6215c51bdd6b30b1fec.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "26e479a1fe517b564c5d20bf8ea18c24b470d53b9d4d6efe40b4cf2d29b84de2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3effb0ac67d1559cbf16941cc3b5f0ce9ba53188550306fc7f5b4192bf9deb37", "generation_seconds_batch": 24.521800426067784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.316492557525635, "genuineness": 4.494794845581055, "identity": 0.3181467652320862, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/26e479a1fe517b564c5d20bf8ea18c24b470d53b9d4d6efe40b4cf2d29b84de2.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "2742e0c752ed579a61b45867b1cc6fee347b04f0076e1d276428032d1eaa1b42", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "79c5df43c9c716c81645b77b8d6461fda0bebe2e5f6c0f8125722322392ab146", "generation_seconds_batch": 18.00894882308785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.300607204437256, "genuineness": 4.400654315948486, "identity": 0.5297080278396606, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2742e0c752ed579a61b45867b1cc6fee347b04f0076e1d276428032d1eaa1b42.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "2756bba8b5ea598094a6e9ea53312773d030ec2bf99086b67da18e5611d855ab", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "463f37a578f865256496bbf0c2cc731bbd0890877f4500145280ec97df2f713a", "generation_seconds_batch": 15.858937766985036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.213237762451172, "genuineness": 4.15588903427124, "identity": 0.6882301568984985, "hypothesis": "ik 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_WARM_reg_high/2756bba8b5ea598094a6e9ea53312773d030ec2bf99086b67da18e5611d855ab.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "27a811e34b5eee53297110a57cd32ceeaa33063723fb2cfbe1fad5c00558311f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f2030053855319f63044a8fb1d7e1baf9a12dd6ef622f7094e6c28bb1d3d74fd", "generation_seconds_batch": 17.229789471020922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.205295085906982, "genuineness": 4.607763767242432, "identity": 0.6164819598197937, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/27a811e34b5eee53297110a57cd32ceeaa33063723fb2cfbe1fad5c00558311f.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "27cc472c9a005117884ad6f62a98d5ca02e87b3d78e33e01109f8c387efc3297", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9a51b50238a2f126756eb1e97877c6d551de5b1c6c05fd7ad03f5fc9bbd52042", "generation_seconds_batch": 17.496656198054552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.332377910614014, "genuineness": 4.507347106933594, "identity": 0.4372682571411133, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/27cc472c9a005117884ad6f62a98d5ca02e87b3d78e33e01109f8c387efc3297.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "280296188302fb22f0bc7886a00b22468fe055cf9daf1cd937e0d2b1f11478c7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "8dc2cd780f78312037be1ce51091bec55e7de474d7c014b6821a91734c9eaa7e", "generation_seconds_batch": 15.892186530050822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.570511341094971, "target_z": 2.171874950595958, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.7591580152511597, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/280296188302fb22f0bc7886a00b22468fe055cf9daf1cd937e0d2b1f11478c7.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "28ed251543b19eb8d6ea985700ad7c7aee4b23391b92d266b84f0ae94ceb77ed", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cc8252f5b8e6d1abc15782923e6b43e90ae4de876c4470de1f35527d1f96bc27", "generation_seconds_batch": 23.350336343050003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.062326908111572, "genuineness": 4.287685871124268, "identity": 0.6200246810913086, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/28ed251543b19eb8d6ea985700ad7c7aee4b23391b92d266b84f0ae94ceb77ed.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "315ee65df6d151e30cb26cdc53def5a834e03dcf8d2d9361dc72c5f61f107020", "generation_seconds_batch": 21.224965500994585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.062326908111572, "genuineness": 4.381826400756836, "identity": 0.6014543771743774, "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_WARM_reg_high/29bec3ae16bfd3c93a38b166264fdb2f776da28e62575f9efb8aafbd1a23d660.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "2a52b980b6f82fa9a3cb1d38478a42f1e1d6ad0fcf7c07bd229d9355225d94ac", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "919b0c6b8c37d2f51c9d328e6130af871a308820f2d1ae36ac420ddc3bee2345", "generation_seconds_batch": 21.03391462098807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.393493175506592, "target_z": 0.8710936553089189, "ce": 6.348263263702393, "genuineness": 4.457139015197754, "identity": 0.4783642292022705, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2a52b980b6f82fa9a3cb1d38478a42f1e1d6ad0fcf7c07bd229d9355225d94ac.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "5f8c74f950d70d029283cf4be3e6da22482cbb62575dd0047ed2b6a470791557", "generation_seconds_batch": 16.757155289989896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 5.927301406860352, "genuineness": 4.375550270080566, "identity": 0.6044397354125977, "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_WARM_reg_high/2ac4242fd97f914ff925f6e4bc35d3005b351dc5362f0eeb9834d8310d117683.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "2add1745823d5de765416154c68366480bf94ee0d328870f3c53712cfc37635e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7696161e4a88a9c09e95199936016a5002323bc27a205ffb6958d66747d7568c", "generation_seconds_batch": 23.91956996207591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 5.998785495758057, "genuineness": 4.21237325668335, "identity": 0.6263601779937744, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2add1745823d5de765416154c68366480bf94ee0d328870f3c53712cfc37635e.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "2afd193d202c72535da8957f49795dea595e710ee0d4941a6f11fe55dd1a99d5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b057d4a693978eb455a3576a5117fedb1fe97f42d3320148e99b2ef0527ef730", "generation_seconds_batch": 24.07497872703243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.5002054572105408, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2afd193d202c72535da8957f49795dea595e710ee0d4941a6f11fe55dd1a99d5.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "2bbc4a03eeaef9238877cb23fde005b232bd1d9d0207ab8f26ba52da78b2e3bd", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "69540818457611c6af25414d3253d9b8f0a9578a35fa21128a856ed73de0141f", "generation_seconds_batch": 16.347157882992178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 5.974957466125488, "genuineness": 4.494794845581055, "identity": 0.6909671425819397, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2bbc4a03eeaef9238877cb23fde005b232bd1d9d0207ab8f26ba52da78b2e3bd.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "2c056d03cfddbf35731fe81bd608c93b87c70c3e4a9439b8afac8afb7f6eff90", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9995cce57817e289fe03fedbba46142b3d364a4d2772e68f0b9e8bb8f6429ecf", "generation_seconds_batch": 23.354315256001428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.300607204437256, "genuineness": 3.8922953605651855, "identity": 0.5321478843688965, "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_WARM_reg_high/2c056d03cfddbf35731fe81bd608c93b87c70c3e4a9439b8afac8afb7f6eff90.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "2c2b82c0ab1fc6f1614e698e086395985979b5d0e399a8a493650982c92ae492", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6b9feb04846290762eca8eff2e522f206af3b656e493dcd9d8383e4c74981ec5", "generation_seconds_batch": 16.345727602019906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.356205940246582, "genuineness": 4.381826400756836, "identity": 0.6753937005996704, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2c2b82c0ab1fc6f1614e698e086395985979b5d0e399a8a493650982c92ae492.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7a25cb750e75a87463910c1d84d929d9195114da5d79f1d795f9697ab7da1a85", "generation_seconds_batch": 17.422137067071162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.4515180587768555, "genuineness": 4.444586753845215, "identity": 0.3821413516998291, "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_WARM_reg_high/2c2bbd678f5c321ed6f491b94824602f66bdc9d09326cfdea4b7bee6e7ce670a.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "2c418c10a68c9f0e1f23a1911353545cd811014092f2abee20e1b12e7f9ae0bd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "37dd252118cf6e13dace3c16990f2c9c50d9d200f5d10056e9255cd0b87ae760", "generation_seconds_batch": 22.961900369962677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 5.974957466125488, "genuineness": 3.7918784618377686, "identity": 0.5508957505226135, "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_WARM_reg_high/2c418c10a68c9f0e1f23a1911353545cd811014092f2abee20e1b12e7f9ae0bd.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "2cd55b7399219e5f7954b235c4ca9e4163e1f8a654d4cfdb7382a6abe596399c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8be72e56d7925adb50509be76f34ffa0b1f44da5d9188b62062dd1b81a3d7fbb", "generation_seconds_batch": 24.134065633057617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.701221764087677, "hypothesis": "He encontrado la cajita. Estaba junto á la ventana vieja. Podemos mirar juntos lo que", "wer": 0.17647058823529413, "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_WARM_reg_high/2cd55b7399219e5f7954b235c4ca9e4163e1f8a654d4cfdb7382a6abe596399c.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "2d53caee6876b64ffa842e3c1be932fb33a872f83fa60a9405107bad543eb613", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "229e2cb0d317864f246ff9464361f6fdcc8fff5ca05243599abb51eb0b06331d", "generation_seconds_batch": 17.515191377024166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.2767791748046875, "genuineness": 4.444586753845215, "identity": 0.650581419467926, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2d53caee6876b64ffa842e3c1be932fb33a872f83fa60a9405107bad543eb613.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "2de9735dc5a00049c3117ef71f04e81dd43d1347548eba0062fecb5de3e82842", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4a6d81ee7162f5217b56a7e2051c99ca2ed2f6c4963bc9c5bc636c5cf8194a14", "generation_seconds_batch": 24.176765793003142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 6.284721851348877, "genuineness": 4.287685871124268, "identity": 0.5394333004951477, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2de9735dc5a00049c3117ef71f04e81dd43d1347548eba0062fecb5de3e82842.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "2e8df297122630a53cc15f5a13e5e466958db7c3eb94408dc32c5af1659b8da6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "73e3345ae633521c5ceaa2ca69b9eb3d10b17464ca7b4ae3ab4d272cd84dd928", "generation_seconds_batch": 21.53973842598498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 5.879645347595215, "genuineness": 3.722842216491699, "identity": 0.6120829582214355, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2e8df297122630a53cc15f5a13e5e466958db7c3eb94408dc32c5af1659b8da6.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "2ee974711c726261f21a9eaaa9d9ca125f3e997ea4375b03a56cb9d2538f54a2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4e2245ceef5deb0d859cb96db4e4061f1bba17d49fbbd4313b2f09d791a7c685", "generation_seconds_batch": 23.426912466995418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.038498878479004, "genuineness": 3.9299514293670654, "identity": 0.45958948135375977, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2ee974711c726261f21a9eaaa9d9ca125f3e997ea4375b03a56cb9d2538f54a2.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "2f2061ccfbf96f6af97dcbbd1d57773b50fcc63af72d4f7931063b1ea1abda0e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f703d76e6b10e537027af34078c0a4800c08f502e8e0d7e61e07400a6269bfaf", "generation_seconds_batch": 23.690823222976178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.09409761428833, "genuineness": 4.369274139404297, "identity": 0.5230051875114441, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2f2061ccfbf96f6af97dcbbd1d57773b50fcc63af72d4f7931063b1ea1abda0e.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "2fa6382e472ffde60cb12b87fbeb27318a2cbc442bc9104a241d17b81133aa0e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "00754e8d881e247d67c48dbaf6be64bbbded8955c477755986d8c85bc1395364", "generation_seconds_batch": 17.059889992000535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089807510375977, "target_z": 1.6406252470202116, "ce": 5.772419452667236, "genuineness": 4.036643981933594, "identity": 0.6924521327018738, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/2fa6382e472ffde60cb12b87fbeb27318a2cbc442bc9104a241d17b81133aa0e.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "2fa9c0798210fb273f4f7b149643636fe40f787861155b3a4a08d8eae752b04a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1a295148eb677415a4089ff5300135ec3e742007333e0aa4354ec8cf4362e316", "generation_seconds_batch": 15.634909820044413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 6.23706579208374, "genuineness": 4.3127899169921875, "identity": 0.7814107537269592, "hypothesis": "De deur 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_WARM_reg_high/2fa9c0798210fb273f4f7b149643636fe40f787861155b3a4a08d8eae752b04a.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "30187800c4c9572d54ac8afe84b555a8fceaa786c1e4ad0c89540609d5acab36", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d122425f6201529fb0198409c97801a1715449da858bb654d42f1533e13d21a3", "generation_seconds_batch": 22.219682855997235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.082737922668457, "target_z": 1.6328122941498235, "ce": 6.006728172302246, "genuineness": 4.243753433227539, "identity": 0.6488655805587769, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/30187800c4c9572d54ac8afe84b555a8fceaa786c1e4ad0c89540609d5acab36.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "3023c8af012e22bad9465f0e89ee1643d4ffa24b1c16bd0e11d2c462a9606dbb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "49918740de5020654c49d1bdfa4cd0d87c5cb85cd00e0476c0c244e28b250929", "generation_seconds_batch": 17.48949614400044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.1020402908325195, "genuineness": 3.9550557136535645, "identity": 0.6074203848838806, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3023c8af012e22bad9465f0e89ee1643d4ffa24b1c16bd0e11d2c462a9606dbb.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "304abb98aac1636e2e64c4c918635bd53c09bff23187b620bd1c33a9ab0f0ab6", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "578633fa772fedd21e1c119c8334d1aa41707e4678494f48b98cd8e529944d17", "generation_seconds_batch": 17.88476680603344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.4853706955909729, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/304abb98aac1636e2e64c4c918635bd53c09bff23187b620bd1c33a9ab0f0ab6.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "3052d56393f0280d2513e295a3b354944461a75d40eb0602d42f5d11d49289de", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "84bd2c140bc147aafe68c0ab3ce200006d267444891c7b0a991eb6bb0db996a6", "generation_seconds_batch": 16.32790992502123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.395919322967529, "genuineness": 4.036643981933594, "identity": 0.5741434097290039, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3052d56393f0280d2513e295a3b354944461a75d40eb0602d42f5d11d49289de.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "309029e75bdd2348d8c5920c0dd8474e212b1107c8e94ae3e5a5952964286d03", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e69b9e13b12735e7aa98bb04290fb17633595a5f8fbdfc76431e71a112a28822", "generation_seconds_batch": 21.929223209037445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.348263263702393, "genuineness": 4.093128204345703, "identity": 0.4419737458229065, "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_WARM_reg_high/309029e75bdd2348d8c5920c0dd8474e212b1107c8e94ae3e5a5952964286d03.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "3096653eab808e2c20a8727fac8e6015901e3b602e8b0cd48ccfb43e9f358e0d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f41e78cd705ba38bf4aedb4838a3f2247c465b4be6dbfcba4b5ea5662d81cc03", "generation_seconds_batch": 17.60515574994497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7716941833496094, "target_z": 1.2890623929579084, "ce": 6.23706579208374, "genuineness": 3.9927117824554443, "identity": 0.39970874786376953, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3096653eab808e2c20a8727fac8e6015901e3b602e8b0cd48ccfb43e9f358e0d.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "309f7beb7f6381e75af3e91b6feff6328329863a5e3e406d0332fafd97e57d92", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fb0dc7194f6d87bf9246e9609f3d7ca2a6173606ba19020acd92ca1f9e47cddb", "generation_seconds_batch": 17.193626957014203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.287744522094727, "target_z": 1.85937527995624, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.36835360527038574, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/309f7beb7f6381e75af3e91b6feff6328329863a5e3e406d0332fafd97e57d92.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2c401d3729d527930818ffe830bc6fefec6fc2d318b2d3a6058e2c143e9189e6", "generation_seconds_batch": 17.50107441598084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941354513168335, "target_z": 1.4765624588299646, "ce": 6.459460258483887, "genuineness": 4.262581825256348, "identity": 0.46291014552116394, "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_WARM_reg_high/314473a397d8c77c6cde3dcada43cbedcca62e052af529615bf4b19bd41f4934.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "31708cb03dedd054f4e8a66ecc41293718d538a09afe0077923fd0db76533d4b", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "aa07843270611d29ecd08d8049c956be74874385a6b73dc8422f83e47846fd2b", "generation_seconds_batch": 15.35720837302506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.3641486167907715, "genuineness": 4.381826400756836, "identity": 0.2782594859600067, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/31708cb03dedd054f4e8a66ecc41293718d538a09afe0077923fd0db76533d4b.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c20846a39825b6537e036a31e4ddda37863a28be8d2c3e7218b7d6fb249c6929", "generation_seconds_batch": 17.171640424989164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0120463371276855, "target_z": 1.554687508234007, "ce": 6.086154937744141, "genuineness": 4.720732688903809, "identity": 0.813940167427063, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je wegge.", "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_WARM_reg_high/317baa1dea53edb7172c54cbed1707f0235670d6ce86231655ca337c40106a47.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "318d88f06776b8671ae92a81da6f65934591104730b003302ec4dc6f2f85e907", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b032a5b8d1f21c6794f6fcb23507ff85ca902fb29caf2c09f31fe4947a41e21a", "generation_seconds_batch": 23.325892319902778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.23706579208374, "genuineness": 4.519899368286133, "identity": 0.8132892847061157, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/318d88f06776b8671ae92a81da6f65934591104730b003302ec4dc6f2f85e907.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "31b90f4c1f1e7b702c4b7d9f737744632ce6d8d37c1ece568d70b0bfcd355685", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1a53b19d92fd39e1d48a1b62010873dac0cb5a17b2f89b7ed59d67e1408badbc", "generation_seconds_batch": 24.14904445607681, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.443265914916992, "target_z": 2.031249703575746, "ce": 5.967014789581299, "genuineness": 4.218649387359619, "identity": 0.5573345422744751, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/31b90f4c1f1e7b702c4b7d9f737744632ce6d8d37c1ece568d70b0bfcd355685.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "328047dfe212cda8181175b0d2457bc5eed23d91ffd2de62535b1b9825848b58", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1759270448ade4bc5cbca7a0307d1f66931d607d719cd1d998ba6ad355a26ab4", "generation_seconds_batch": 24.258861553040333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.1894097328186035, "genuineness": 3.986435890197754, "identity": 0.12151002138853073, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/328047dfe212cda8181175b0d2457bc5eed23d91ffd2de62535b1b9825848b58.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "32d06d77e3b2b293748bc336d1b79da23fa08eadfee73bf291ff6dcbc8625ca2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7654b4383c2e65ecb5e992a670490602138d1c806bc60e61a35461f61858dd15", "generation_seconds_batch": 15.331698611029424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.443574905395508, "genuineness": 4.093128204345703, "identity": 0.5807400345802307, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschießen.", "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_WARM_reg_high/32d06d77e3b2b293748bc336d1b79da23fa08eadfee73bf291ff6dcbc8625ca2.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "32da2c61154c5cc07bc8aeef2f5d14ddc59182f562767f19efa5b401160ab477", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f6825e112650fe0f1d0163dc67ba6a47bbf8e3bfa204e7a6f2d802fef99f745f", "generation_seconds_batch": 23.967351846979, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.070269584655762, "genuineness": 4.406930446624756, "identity": 0.7863218784332275, "hypothesis": "La porte 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_WARM_reg_high/32da2c61154c5cc07bc8aeef2f5d14ddc59182f562767f19efa5b401160ab477.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f6f975bbd6eca3e700d2ec51ace9b172b228ce7c1bc88e1ea667fb03ec380929", "generation_seconds_batch": 18.061762645025738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.217052459716797, "target_z": 1.7812499670639719, "ce": 6.165581703186035, "genuineness": 4.369274139404297, "identity": 0.7562160491943359, "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": 0.120000000000001, "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_WARM_reg_high/33078f6278fc7f9eea574027b2f45a5f07e06059176577e01448eb02f6ae41e1.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "338594ea43778aa7632c0a9639fa1c0ac42323c089bd1ed3bef7f42781f01167", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b6d9f408e99d9d418411bd7ed78042fe46386f19c4c87819ea80b97e5eee394", "generation_seconds_batch": 23.864452486042865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.146360874176025, "target_z": 1.7031251811481551, "ce": 5.847874641418457, "genuineness": 3.873466968536377, "identity": 0.3656643033027649, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lenta.", "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_WARM_reg_high/338594ea43778aa7632c0a9639fa1c0ac42323c089bd1ed3bef7f42781f01167.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a7f3cb4fc5447e848ccd2261375fb3e725d63d7b91d1d1d495783cdf79643e17", "generation_seconds_batch": 24.29494415107183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 6.078212261199951, "genuineness": 3.9425034523010254, "identity": 0.8498762249946594, "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_WARM_reg_high/33aca44d832266c05db5caf317062e6d735de11cad55b9ec83105dfba223295c.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "33d532f109982f3125e68d15f89039e5491504dbc6ffaed09cd421a58ed113ab", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "c8c030531883e9b84cb2cabb1277af7c8d86e2de085d2fcab8006f188b8f7944", "generation_seconds_batch": 22.257453154074028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.7352933883666992, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/33d532f109982f3125e68d15f89039e5491504dbc6ffaed09cd421a58ed113ab.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very warm vocal delivery. 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": "237a6d1d6175294014db058ab47163b416fecb250db400cefca52e7fd9d15940", "generation_seconds_batch": 17.560704670031555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2415058612823486, "target_z": 0.7031249176599295, "ce": 6.181467056274414, "genuineness": 4.13078498840332, "identity": 0.6354613304138184, "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_WARM_reg_high/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414", "bucket": "vn_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very warm vocal delivery. 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": "186cb8a646da590d8fd52e012428bde2e804bdde1581a9788abd276154e3c2cf", "generation_seconds_batch": 17.802157877013087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.41180419921875, "genuineness": 4.030367851257324, "identity": 0.26380473375320435, "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_WARM_reg_high/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "34827f5fd7d93d62f474d0fdc322b53474187750a9f8cee52e882da067d399d8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a131bf2829f44382cc2883d16999deb91ba4ddfcb5fdd36ad3e5c7838fab95b4", "generation_seconds_batch": 24.91056120104622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.443265914916992, "target_z": 2.031249703575746, "ce": 5.935244083404541, "genuineness": 4.319066047668457, "identity": 0.6289240121841431, "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_WARM_reg_high/34827f5fd7d93d62f474d0fdc322b53474187750a9f8cee52e882da067d399d8.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d9f29fb014ebf422d2d8b79118bcaf4b8c4a46a3e6ae04debe2844931d8a2094", "generation_seconds_batch": 15.340972983976826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.165581703186035, "genuineness": 4.494794845581055, "identity": 0.7232900857925415, "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_WARM_reg_high/348bce606c41aac89cdcf820462f40ffc028b6c0fd00aa96c2eb592f7ea76c60.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "34a54af028975c8ea973d46aa4a3b542898b442a711e81af3ad840c6cd9a8028", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "070194fe2907c7fa70d1dc914b80645e6d947dcabb97a68475dde38f8715fa0b", "generation_seconds_batch": 17.08630894299131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.348263263702393, "genuineness": 4.325342178344727, "identity": 0.42286691069602966, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/34a54af028975c8ea973d46aa4a3b542898b442a711e81af3ad840c6cd9a8028.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "34a6e83651ae28bcc734faf8c0e8f4aae8b1c44bff0129732aed64e7428626d5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "08656e4cb243c1e4d9a525466ff16b8206326fd98da02e7ad0081a525044b1a1", "generation_seconds_batch": 23.497221046010964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 6.117925643920898, "genuineness": 4.193545341491699, "identity": 0.7805290818214417, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/34a6e83651ae28bcc734faf8c0e8f4aae8b1c44bff0129732aed64e7428626d5.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "34aaadd618dd118c489a6b4c3233b38c9bca5631cb482eb7e39e03b4e5be5f31", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3d1fe4b53335fc7f0d1f72b099c93b02e0dd5326d6012c205936ac1a0effe2cc", "generation_seconds_batch": 22.107812745962292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1319336891174316, "target_z": 0.5820312623510105, "ce": 5.434855937957764, "genuineness": 4.18099308013916, "identity": 0.7986329197883606, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar staat.", "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_WARM_reg_high/34aaadd618dd118c489a6b4c3233b38c9bca5631cb482eb7e39e03b4e5be5f31.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "36419ea1ee6428dd90e99a43176f3dcc3f0e7b163f651feb28ee449b2da4549e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b18cda5e500507082d7317eb454b2f473ae7650e60680e6a720fa4d8d6b3cb89", "generation_seconds_batch": 24.399975895998068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8141093254089355, "target_z": 1.335937475297979, "ce": 6.324435234069824, "genuineness": 4.381826400756836, "identity": 0.6109398007392883, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/36419ea1ee6428dd90e99a43176f3dcc3f0e7b163f651feb28ee449b2da4549e.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "36ad4387ae852dd967807dd8096825387a41139aa6e3c572a9ecee994311dea5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "5aedea432a4880ef8ccb34e9b040484c6b7811a0b464a627292dd5d3fded335f", "generation_seconds_batch": 24.067220276920125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 5.764476776123047, "genuineness": 3.8546390533447266, "identity": 0.6826019287109375, "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_WARM_reg_high/36ad4387ae852dd967807dd8096825387a41139aa6e3c572a9ecee994311dea5.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "36d5d93e589b9fa5e96da13b8712241513d281888dc65dc56f3d315b9bebaf71", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "090f12480beff2c63a9fe9d5c7e77cc291fe5eb46d1ba93f3903cb6143836ff2", "generation_seconds_batch": 29.340077362023294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.427689552307129, "genuineness": 4.337893962860107, "identity": 0.581430196762085, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "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_WARM_reg_high/36d5d93e589b9fa5e96da13b8712241513d281888dc65dc56f3d315b9bebaf71.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "37047e35353641e578872d296823a8b4b2a015208613c4b85872364ee28305bf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d0a8b5f75c7c53ec3d523f26105fc43eb305fe39e3cebcd14a56175b610f72e1", "generation_seconds_batch": 22.612859509070404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.6742890477180481, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/37047e35353641e578872d296823a8b4b2a015208613c4b85872364ee28305bf.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "37b71543aa7a47d623677de0658ce8d24409e3f23d4e4be972748cda45f4a122", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a902497ec04158bd3eb7fdccbca98ef0807f28d22ee804cf8b24bf9d2a81601f", "generation_seconds_batch": 17.515191377024166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.340320587158203, "genuineness": 4.394378662109375, "identity": 0.5122354030609131, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/37b71543aa7a47d623677de0658ce8d24409e3f23d4e4be972748cda45f4a122.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "37c7fa19f0edec14c9c4b22c076e75105e5273b74043ab50fd4f0af6459d17fa", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b85e401562c71b837e9f1449a6f45ddfc32b520a1f6543bad96cab731720865c", "generation_seconds_batch": 22.77544435893651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.078212261199951, "genuineness": 4.162164688110352, "identity": 0.6553248167037964, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/37c7fa19f0edec14c9c4b22c076e75105e5273b74043ab50fd4f0af6459d17fa.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "38112a71c496ea7bbbcc677a6c1e8e32d27d629550d23855dc061df0afd43a33", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "62778ac0aafabce4b71f18050601b7bcc8ee7daa19d06b8385c2ee18fc74b0e6", "generation_seconds_batch": 21.136285643093288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.332377910614014, "genuineness": 4.168440818786621, "identity": 0.25066250562667847, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/38112a71c496ea7bbbcc677a6c1e8e32d27d629550d23855dc061df0afd43a33.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "382fcf47db82d240d20f264b0dbe11bc9cf4f960a3c500544b16e8aba11b24e8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "731925c3b95eb6dc96a64804a5831751b4bf8f9d5359ecf77f24e22886a8d0b1", "generation_seconds_batch": 25.204812011914328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.221180438995361, "genuineness": 4.3504462242126465, "identity": 0.6483277082443237, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/382fcf47db82d240d20f264b0dbe11bc9cf4f960a3c500544b16e8aba11b24e8.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "384ec87325277029d96a6926a8e13f34b9429f869d222d056ee55dc54bcf409e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "23c456394e5be5867332bcd3a5e12b056e9973b565aa02439dcee8f4cdc4008f", "generation_seconds_batch": 17.588252476998605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 5.903473377227783, "genuineness": 4.65797233581543, "identity": 0.7056551575660706, "hypothesis": "J'ai trouvé la 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_WARM_reg_high/384ec87325277029d96a6926a8e13f34b9429f869d222d056ee55dc54bcf409e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "78b1ad7d1c883b9d2a3963fbcd45d602bea95aaa8f2ef9a72b2eb9cc8e65eb19", "generation_seconds_batch": 16.7208095099777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.53094482421875, "genuineness": 4.287685871124268, "identity": 0.350102037191391, "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_WARM_reg_high/3852d05e7618fa390ab8ec1b13b82ed045a04c25b2b1fdfdc25575807f1970a3.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "386fd15d2ac5266f5cef6b6633b6d3413ed97eb040db01b3699347e517aaf3af", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4825847336e5691ef8bd9661320a0b1007aea47e54c9562f039682c4bf87cf83", "generation_seconds_batch": 22.750352279050276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.139291763305664, "target_z": 1.6953127552542187, "ce": 6.038498878479004, "genuineness": 4.030367851257324, "identity": 0.527360200881958, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás abriendo.", "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_WARM_reg_high/386fd15d2ac5266f5cef6b6633b6d3413ed97eb040db01b3699347e517aaf3af.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "3905827e7dae440afef947720e889fe76e7afb171daefb06d7a03c72c5937e1c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "7fb1287b500e1c81c02f69e922a459c50111b13df145c9c3c3263e877309f240", "generation_seconds_batch": 22.03417051397264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.3641486167907715, "genuineness": 3.8671910762786865, "identity": 0.32391873002052307, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3905827e7dae440afef947720e889fe76e7afb171daefb06d7a03c72c5937e1c.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "394c23121f6727e08ba5dd68d80e73c2e2e55d3bb1618d05226fb0839be30dfa", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fe7edd6f789a828b730587511c09850409c055202168cef08dd17f74ff0521f4", "generation_seconds_batch": 17.02996214001905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 5.95112943649292, "genuineness": 3.835810661315918, "identity": 0.4842608571052551, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/394c23121f6727e08ba5dd68d80e73c2e2e55d3bb1618d05226fb0839be30dfa.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "3983ff1d0c548d42a55d70751d53fb1b936968a4b5ddf8c6eba7777a5ca9ff2d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ad9d1b919199a34029f790e3f70b2a5a387d71ac48a826e20f53e882c75629d5", "generation_seconds_batch": 22.48642150394153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.300607204437256, "genuineness": 4.325342178344727, "identity": 0.5135402083396912, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3983ff1d0c548d42a55d70751d53fb1b936968a4b5ddf8c6eba7777a5ca9ff2d.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "3995cb85308812d43b0d9e919adc59fb4af5d107c6d41ad9286de60da56c7418", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "956639eabcfe9073d53e3ee929df342c67fd380c9fe825827e60e3a1cf6555e7", "generation_seconds_batch": 16.62254815909546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.078212261199951, "genuineness": 4.394378662109375, "identity": 0.5465186834335327, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3995cb85308812d43b0d9e919adc59fb4af5d107c6d41ad9286de60da56c7418.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "39b8c2d35dd6c15d2cf8f6551376a19f6935f122d65cb3077406c598a48995a4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1aefc95d37b494fa903995d55620c1a657d8bf63a89aecf1d66146bc0b95b93e", "generation_seconds_batch": 23.966446034959517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 6.38797664642334, "genuineness": 4.457139015197754, "identity": 0.6069109439849854, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/39b8c2d35dd6c15d2cf8f6551376a19f6935f122d65cb3077406c598a48995a4.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "3a3b01e7d3e296d9bfe1f09bf55fa66bc5a3633e35d79023883ed9147ed2f3fa", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a70c02a1b1a738e387bb55fa6ed02223c50dbb43883c094c30313a3725b5413a", "generation_seconds_batch": 16.675720482016914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.006728172302246, "genuineness": 4.3881025314331055, "identity": 0.4913179874420166, "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_WARM_reg_high/3a3b01e7d3e296d9bfe1f09bf55fa66bc5a3633e35d79023883ed9147ed2f3fa.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "20f3e69520d1f4eb6c2f5836803a63979ad71ae8f5c06579244b999113419acb", "generation_seconds_batch": 24.08911552594509, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.701002597808838, "target_z": 1.2109376070420916, "ce": 6.229123115539551, "genuineness": 3.7165660858154297, "identity": 0.5580345392227173, "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_WARM_reg_high/3a5e43149a76290e684da0425e4410fbe973aee376dfd87f45407085bf3f72c0.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "3a68d2ae95ab06dc9606debb0ec36ff37142fc294db446cb44dadc427ed735b9", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8a7f96eed727f12f311c615e54e2a3957461739a22ba9b84dfc9a61dfb6f7d3b", "generation_seconds_batch": 17.386325592990033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.284721851348877, "genuineness": 4.331617832183838, "identity": 0.7039036750793457, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3a68d2ae95ab06dc9606debb0ec36ff37142fc294db446cb44dadc427ed735b9.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "3a78e9a5f4e6ce0a25b54f4bed6e990f762d13742aeb68805187124c872ff276", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "e9e71f225592885e7c37d22674cb704f6b8e74db8a784dbc3a396091bd0127e7", "generation_seconds_batch": 18.909976088092662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 6.372091293334961, "genuineness": 4.068024158477783, "identity": 0.3291853666305542, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3a78e9a5f4e6ce0a25b54f4bed6e990f762d13742aeb68805187124c872ff276.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "3a96f4a44e18acbac3d50a2b23db41d75cdd48fc84e0ef17ee31dfa7d07feebe", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c084310858d55dc97e541c3b5c91e8dd3f8b8d069c99fe1ba456cef01fb0bac7", "generation_seconds_batch": 17.534848776995204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.3641486167907715, "genuineness": 4.20609712600708, "identity": 0.6038574576377869, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3a96f4a44e18acbac3d50a2b23db41d75cdd48fc84e0ef17ee31dfa7d07feebe.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "3ab02ad22c16fa1ee4fe69f089210fa5d1ad0375b87387c62f1a00fed8b21a7e", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "89697ac726a5f6488d6f59337e79741f70cdc34049bd9a5b94b6f854622afbe3", "generation_seconds_batch": 17.776912480941974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0895185470581055, "target_z": 0.53515618001094, "ce": 6.006728172302246, "genuineness": 4.2249250411987305, "identity": 0.53816819190979, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3ab02ad22c16fa1ee4fe69f089210fa5d1ad0375b87387c62f1a00fed8b21a7e.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "3ab517472665ada4437024423b92f5911ae5a95739a057b85a67c3f7dff18ad2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "4a399f856b190de51eb33c61007ddb2e5118c6e8d7175f2ba60cb80de88bfa72", "generation_seconds_batch": 16.929879443952814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2311906814575195, "target_z": 1.7968748188518449, "ce": 6.197352409362793, "genuineness": 4.457139015197754, "identity": 0.7905704975128174, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3ab517472665ada4437024423b92f5911ae5a95739a057b85a67c3f7dff18ad2.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "3abe9eb5e03acf15b59e1d6b49630f543304b5510e210fb6c811e42841db3d4f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "a1bc8db4eb5c6fdd5340eb15d5a98aa6c2b41899d08c104d58df3fad42b290eb", "generation_seconds_batch": 22.676408263971098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.6810686588287354, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3abe9eb5e03acf15b59e1d6b49630f543304b5510e210fb6c811e42841db3d4f.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "3acd6e3218419aa510a8fdb7377554fe44ca8c163ff67353c06c6eec940a2d4f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "270d68311e6068fecf3cbbd008b1c82f2c88f1d023ac84e950c4bab5b62a0bd8", "generation_seconds_batch": 23.870798907941207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 5.617537498474121, "genuineness": 4.65797233581543, "identity": 0.6689337491989136, "hypothesis": "I trouved the petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.5263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3acd6e3218419aa510a8fdb7377554fe44ca8c163ff67353c06c6eec940a2d4f.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "3aff136d5adec239078471bbc09ce0b89986928344cc3b4939da385115280a3e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "99cb0c309626282522d5bd77014b5a7801d477c6843a422e68940b739203101e", "generation_seconds_batch": 23.870798907941207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8706626892089844, "target_z": 1.3984374094259224, "ce": 5.561938762664795, "genuineness": 4.13078498840332, "identity": 0.7728794813156128, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3aff136d5adec239078471bbc09ce0b89986928344cc3b4939da385115280a3e.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "3affc630280a6c29347d79dcea5cb35e54a79eede2ef5d05edff1c7bdf9a5460", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f0d4dba3ff7a4494d5e5fa703425bee4a23d4440b4ae6a19fe7fd127625261e2", "generation_seconds_batch": 24.14904445607681, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.068599700927734, "target_z": 1.6171874423619506, "ce": 6.038498878479004, "genuineness": 4.093128204345703, "identity": 0.7146139144897461, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je knapt.", "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_WARM_reg_high/3affc630280a6c29347d79dcea5cb35e54a79eede2ef5d05edff1c7bdf9a5460.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "3b0b61b24cdd6ce39e021af5edcd9868bc21fc1591fdd778aa00a717760c7394", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "01a5ea8b61e1429a0953e6e2c4ec12284e2988a5f0744bf0d19b415713c9e5f9", "generation_seconds_batch": 17.74324284994509, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.268836498260498, "genuineness": 4.124508857727051, "identity": 0.7730070948600769, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3b0b61b24cdd6ce39e021af5edcd9868bc21fc1591fdd778aa00a717760c7394.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "3b124de3fbaa5c69cb5438a8391af7487c6bbe5d12466f26052a6805a5081d58", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7f329c686190e8aa403ba91ffcfe850eafeff62d1fb50301b0e66aa3b0a5d901", "generation_seconds_batch": 23.690823222976178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 6.181467056274414, "genuineness": 4.068024158477783, "identity": 0.7666164040565491, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3b124de3fbaa5c69cb5438a8391af7487c6bbe5d12466f26052a6805a5081d58.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6d21053ca0cfc0949c908bed6d9706bf7567870e30d0976bb0cc110d78280ced", "generation_seconds_batch": 21.21823633206077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9696311950683594, "target_z": 1.5078124258939365, "ce": 6.205295085906982, "genuineness": 4.024092197418213, "identity": 0.6641250848770142, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego, junto a la ventana, llévatelas cuando te vayan.", "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_WARM_reg_high/3b7a1e68ee692f8235e05225168335ad80b7a0e8d4da1f018d3e7555e1805d6c.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "3b84442f49d371314b21897c5c3d4d0ae34e7f27a81881e45dd96196c2fecfaf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "cef0360a1bde79d505475a4d0886e1012893c9491985d63c9464db6f888cc607", "generation_seconds_batch": 23.777553699095733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174637317657471, "target_z": 1.7343748847239013, "ce": 5.974957466125488, "genuineness": 3.9801597595214844, "identity": 0.6139532327651978, "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_WARM_reg_high/3b84442f49d371314b21897c5c3d4d0ae34e7f27a81881e45dd96196c2fecfaf.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "3bd625fc7e1d0dafdb5b10834d7767b8299b5b1837b26e9701c87ebd1f4ab545", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "6089f81a748397c57dffbb727e535e00a1e4bd6f1bd2d04ff32c43fcd6b0b565", "generation_seconds_batch": 17.667729720007628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.419747352600098, "genuineness": 3.8044304847717285, "identity": 0.5024506449699402, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3bd625fc7e1d0dafdb5b10834d7767b8299b5b1837b26e9701c87ebd1f4ab545.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "3c3e49f48d1dd3a996054437ac422bad86e448f18d4c18d1794782af84509cf4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "75d4ce5571fef4ba2874fce16868fd5cd1a88c2e190d83def85ebaac7e22e93e", "generation_seconds_batch": 22.642339496989734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.197352409362793, "genuineness": 4.545003414154053, "identity": 0.8159286975860596, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3c3e49f48d1dd3a996054437ac422bad86e448f18d4c18d1794782af84509cf4.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very warm vocal delivery. 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": "2fedd30de379973f4e120f283fb99fca702c3ebb36b2319f300c0c8b0d5cb419", "generation_seconds_batch": 17.434214334003627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.252951145172119, "genuineness": 4.21237325668335, "identity": 0.7584497928619385, "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_WARM_reg_high/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "3c70590a9cfac23fd95260ca68c5b178745ecd6e7b29fa68a2b79f7a42c2183e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "6a74c4895efafa2d87d73556bfbb92e0044087c797855d460ded4be75fdc0c07", "generation_seconds_batch": 23.982555918977596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.149696350097656, "genuineness": 4.293961524963379, "identity": 0.22149863839149475, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3c70590a9cfac23fd95260ca68c5b178745ecd6e7b29fa68a2b79f7a42c2183e.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "3c9f8b218bfe6322717219e8abf15a7ea13bb1f9087d3f2340e3f9d71814f0b2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "05ebc9a0168ffd3ce6734bfc2704f423335e28d3cd4c182108bddb3d086a7066", "generation_seconds_batch": 23.37862551410217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.372574329376221, "target_z": 1.9531249176599295, "ce": 5.8875885009765625, "genuineness": 4.281409740447998, "identity": 0.7629215717315674, "hypothesis": "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_WARM_reg_high/3c9f8b218bfe6322717219e8abf15a7ea13bb1f9087d3f2340e3f9d71814f0b2.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "3ccda09a491e2709f3aefeb86f7eb0e7a9d1e615016825088046819d2f25b667", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c6d7f7e22b7da9addbb44246a2b638bf0df5d15ee6fbafff811be2b5fcb942d8", "generation_seconds_batch": 22.197307275957428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.2767791748046875, "genuineness": 4.20609712600708, "identity": 0.6853832602500916, "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_WARM_reg_high/3ccda09a491e2709f3aefeb86f7eb0e7a9d1e615016825088046819d2f25b667.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "3cda83c22a6372236de851e588575d3842ed22b5362da240a2d7157579fbf5c7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c6ceec70724f86b14a80a438db26a1980b9a9db9ae0bd44aef74834854a351f5", "generation_seconds_batch": 22.855786213069223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.300607204437256, "genuineness": 4.519899368286133, "identity": 0.8100181818008423, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3cda83c22a6372236de851e588575d3842ed22b5362da240a2d7157579fbf5c7.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "3f1e636c39edf3021a1051095207c450e748efe7f6f1bf64c997287b3c3eac3d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3727eab9433f1ea70bfdd64bef8eb5af5f536acbd9874a638621d338028dd027", "generation_seconds_batch": 16.84673763893079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3192667961120605, "target_z": 0.7890623929579083, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.4724084138870239, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3f1e636c39edf3021a1051095207c450e748efe7f6f1bf64c997287b3c3eac3d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "3f734c43192d4922ff712128feaa17aa44d966a78dbbaa872a043de1f066fd49", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d7e840c21466069a882ab8abf167e9b22d85e37b11474663f031153fd9d7cd09", "generation_seconds_batch": 22.257453154074028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.070269584655762, "genuineness": 4.331617832183838, "identity": 0.5706756114959717, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3f734c43192d4922ff712128feaa17aa44d966a78dbbaa872a043de1f066fd49.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "3f82459ff4e5b38b61f12b6d4d343f7e57650b8b808644532234546fb487c5fd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b2f8dadd39cac80995d16ef2432a6beb2cbfc0bd2a4e88dc643d2e16cc34ff89", "generation_seconds_batch": 24.00880176492501, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.570511341094971, "target_z": 2.171874950595958, "ce": 6.141753673553467, "genuineness": 4.419482707977295, "identity": 0.6805803775787354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3f82459ff4e5b38b61f12b6d4d343f7e57650b8b808644532234546fb487c5fd.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ef853e2ddcc49bba9d251a059dbed8eb653b80b8f9d25bb08e750c0a9d1c1de5", "generation_seconds_batch": 17.795576908043586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.300607204437256, "genuineness": 4.457139015197754, "identity": 0.7286266088485718, "hypothesis": "Jemand hat dem kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "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_WARM_reg_high/3feabf3dca111faea220f644e43c1d9e14d6e044cbcbce79d48b0de1245c789e.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "3ffe57ee8811fe0bd6c00d1ee9856e0521a40da2568d75ff1a39768eafe45d5b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "25c0a2dc96f3011b554936f456f1758b69da439cb5b43e2e4e69f9adf3726176", "generation_seconds_batch": 21.997723033069633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.173524379730225, "genuineness": 4.218649387359619, "identity": 0.8313023447990417, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/3ffe57ee8811fe0bd6c00d1ee9856e0521a40da2568d75ff1a39768eafe45d5b.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "40b427c19b25a9c216fc5f2b4c5d8310ee5aae18da5aa27bbc24ecacc4439642", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "675e91aef8891b31b7999059587ceb52b0c99c01850de60cd3b62ba9dac0b0b4", "generation_seconds_batch": 15.331698611029424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.332377910614014, "genuineness": 4.256305694580078, "identity": 0.5834336280822754, "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_WARM_reg_high/40b427c19b25a9c216fc5f2b4c5d8310ee5aae18da5aa27bbc24ecacc4439642.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "41176a582ca4f609329bde2786b06a0a4f94978fe654cb91d0684904f15cd2f3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "60334c2258bef979e50c8009de2fe5cba2e48ad252832c09aa13f62f7ec20f2a", "generation_seconds_batch": 21.084210560889915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.316492557525635, "genuineness": 4.281409740447998, "identity": 0.42266103625297546, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/41176a582ca4f609329bde2786b06a0a4f94978fe654cb91d0684904f15cd2f3.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "41616f818059dc0ff481ea8d7a0f30a5e8ab8d278cfb924b1f55018d49c939d2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "235779ca052044b097e63a7417947f6a92eb03e494fd92640d7c62c81f9c1c76", "generation_seconds_batch": 22.612859509070404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.7759106159210205, "hypothesis": "De deur 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_WARM_reg_high/41616f818059dc0ff481ea8d7a0f30a5e8ab8d278cfb924b1f55018d49c939d2.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "417cb142d2d97bec08f1e61c5b8f7df0186aff645d0504d2e9ebb02646d37bf8", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9a2d3e711edb55ea9882bf39b504b07bf588baa5dff6b147c54c1c71267505f3", "generation_seconds_batch": 18.06516317697242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.09409761428833, "genuineness": 4.300237655639648, "identity": 0.6820693016052246, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/417cb142d2d97bec08f1e61c5b8f7df0186aff645d0504d2e9ebb02646d37bf8.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "41a1a925686c0efbc9717c429cade9351a0447b9a10480f246bd191a158f1bf9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "aa810c1b81798c2e847ae32135498f7eebf7a8c2ab0f5d0d1199e467c7c60f7e", "generation_seconds_batch": 23.6397933709668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743417739868164, "target_z": 1.2578126893821622, "ce": 6.1894097328186035, "genuineness": 4.344170093536377, "identity": 0.7697553634643555, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/41a1a925686c0efbc9717c429cade9351a0447b9a10480f246bd191a158f1bf9.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "41c90986ecff94ed6501f57afb1ee9411cd2d8e4f6bb9e0951f37dd45220dc09", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7b1cae5d2fe534d31e600bb2109c75ab159cc54be751a20378f288554e28dfb1", "generation_seconds_batch": 18.50474062305875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.1894097328186035, "genuineness": 4.331617832183838, "identity": 0.6489009857177734, "hypothesis": "Die Tür 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_WARM_reg_high/41c90986ecff94ed6501f57afb1ee9411cd2d8e4f6bb9e0951f37dd45220dc09.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "420f21b952ff4702ec6ab7599c9ebd2343460bedcdab8997f12f168473c4577f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9fa2e5171a13e1d56c7d66b4c85dd9c8dbff2e17d8853cabd3c87e08b459bc60", "generation_seconds_batch": 17.48949614400044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.125868320465088, "genuineness": 4.394378662109375, "identity": 0.6395505666732788, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/420f21b952ff4702ec6ab7599c9ebd2343460bedcdab8997f12f168473c4577f.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "428c4c719b1262836c11b14c08b1741caf30fa9b171af641e0c78b1fe7d125bf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "faa9d89c6a11b2dc5eb635d9f7d24e0c3387b9e0a8628956d2065359207e9f70", "generation_seconds_batch": 26.514590451028198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.188776016235352, "target_z": 1.7500002634882257, "ce": 5.871703147888184, "genuineness": 3.7793264389038086, "identity": 0.5172915458679199, "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_WARM_reg_high/428c4c719b1262836c11b14c08b1741caf30fa9b171af641e0c78b1fe7d125bf.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(very warm vocal delivery. 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": "8283a8f06c35382467b102e67adec9cbbf359c3a8998aa0259665d718eca7de3", "generation_seconds_batch": 17.560704670031555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3192667961120605, "target_z": 0.7890623929579083, "ce": 6.078212261199951, "genuineness": 4.319066047668457, "identity": 0.8205075860023499, "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_WARM_reg_high/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "437b225a3dae2f33c0be2d36a25b7a1d0a528442b155caf9310d1f3755f31839", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c3cd9e920d823218a08f15269d3ce629f623b7b15519893d59d9bc7b7dcde65a", "generation_seconds_batch": 21.18986849905923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.7155187129974365, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/437b225a3dae2f33c0be2d36a25b7a1d0a528442b155caf9310d1f3755f31839.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "43a25408af9d3c0af462c123c31f64bf0235cf52ec435478f93a224a8e4863b7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4221ddf2d8244b6882dc671b6455d000fb1c2b965c810ae4e5fb21b0345566d5", "generation_seconds_batch": 21.437781027983874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.3641486167907715, "genuineness": 4.055472373962402, "identity": 0.30897852778434753, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/43a25408af9d3c0af462c123c31f64bf0235cf52ec435478f93a224a8e4863b7.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "43c24919b339e49b994a2d818320754c9f479130e35d20d80d76a9eadff268e4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "dfdab2c21e4bf699d0c94c47fef306f8e0eb1f47eba1ebbbef29ae646157e07b", "generation_seconds_batch": 24.91056120104622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.556373119354248, "target_z": 2.156250098808085, "ce": 6.173524379730225, "genuineness": 4.632867813110352, "identity": 0.660500168800354, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que había.", "wer": 0.11764705882352941, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/43c24919b339e49b994a2d818320754c9f479130e35d20d80d76a9eadff268e4.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "43c3d113d84b66952d7d6146c2427e4322c8ba4a257a5b1953c64b90b413417c", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3c2fc1bf5d881f55941dac79c31ad02a27c092f4f20c6540513f30483cf9eac4", "generation_seconds_batch": 15.254763682023622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0046885013580322, "target_z": 0.44140627881902467, "ce": 6.1894097328186035, "genuineness": 4.21237325668335, "identity": 0.5596189498901367, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/43c3d113d84b66952d7d6146c2427e4322c8ba4a257a5b1953c64b90b413417c.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very warm vocal delivery. 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": "c3d91a76a3f31366104d9d86edebb7cd8ce1b64c996021406ea45c29ec2eb0b1", "generation_seconds_batch": 15.486682903952897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.24500846862793, "genuineness": 4.519899368286133, "identity": 0.7599644660949707, "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_WARM_reg_high/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "447b4177fd9a0adb30b0caf5e973e9eec51efa188a9f87cc24c2639314d42355", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d138715c1fb23e2fcbe0a87270579f4d7f0d3aad5905db7bfafb127d9bfd12b5", "generation_seconds_batch": 25.33071229699999, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.393781661987305, "target_z": 1.976562195341739, "ce": 5.990842819213867, "genuineness": 3.923675537109375, "identity": 0.4264217019081116, "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_WARM_reg_high/447b4177fd9a0adb30b0caf5e973e9eec51efa188a9f87cc24c2639314d42355.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "4501d92f17fd604e3d5cb6c73b6f5bbf815d2f2f228646dad1fbfb4a2a7fbe67", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b128cf2b5529203862309c23d94abd547807f3aab60b78caf3fb2249a2af4e29", "generation_seconds_batch": 19.019884507986717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1637449264526367, "target_z": 0.6171874423619507, "ce": 6.316492557525635, "genuineness": 4.174716949462891, "identity": 0.24652083218097687, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4501d92f17fd604e3d5cb6c73b6f5bbf815d2f2f228646dad1fbfb4a2a7fbe67.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "4543fc485841c15e995d75fc9d19a6470c23e47170775a3f2de5ca05e4423688", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3d1afbf05c8a11e9ff6dbd493f7bf86aa184494a8ad2f4b556b8e99433f33612", "generation_seconds_batch": 17.883928347961046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.488927125930786, "target_z": 0.9765624588299647, "ce": 6.1894097328186035, "genuineness": 4.595211505889893, "identity": 0.6612939834594727, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4543fc485841c15e995d75fc9d19a6470c23e47170775a3f2de5ca05e4423688.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "4557e27d743bde6350656cc5e1df94d7a392db0cbeb34694e3a36c0473fc0fc4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "dda27f28e426f3e0f63c7f63549da18ada723dc6fab9b515237347dafedc4946", "generation_seconds_batch": 24.791064786957577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.23706579208374, "genuineness": 4.024092197418213, "identity": 0.16242575645446777, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4557e27d743bde6350656cc5e1df94d7a392db0cbeb34694e3a36c0473fc0fc4.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "4559eb89e2aed4d908852f196bd454e452cf722c2492c4999d776dc7d0c5db99", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "21883cbc2496539d2a5212f2ae22d623ab569e1fc210de9f078703509a86137f", "generation_seconds_batch": 23.29593155893963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8141093254089355, "target_z": 1.335937475297979, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.7177851796150208, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4559eb89e2aed4d908852f196bd454e452cf722c2492c4999d776dc7d0c5db99.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "457997508dce34669282eba03b2b365164daa8e6618fc9b3aeb7c95b25549fa5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8853e890f9bdb53e52a8fb6ed93724382210d5bf104091eca302204e1c0da703", "generation_seconds_batch": 16.32337344996631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.046441555023193, "genuineness": 4.3504462242126465, "identity": 0.6372997164726257, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/457997508dce34669282eba03b2b365164daa8e6618fc9b3aeb7c95b25549fa5.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very warm vocal delivery. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ef8846c1c270c13a21427fad2a1d7f679a8c8fd1aa718392730a1884751b8ce8", "generation_seconds_batch": 17.999235155060887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3440089225769043, "target_z": 0.8164061470749118, "ce": 6.332377910614014, "genuineness": 4.344170093536377, "identity": 0.47806838154792786, "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_WARM_reg_high/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "467a925c3cce13161bb798fcea482a4bcb118ecd1dbb0b9e73f2c3a47e4e588b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1dfef836a971b37f0dd865e3eaf221fd96291d78406f70542065066f3158118a", "generation_seconds_batch": 22.113278625998646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.146360874176025, "target_z": 1.7031251811481551, "ce": 6.268836498260498, "genuineness": 4.243753433227539, "identity": 0.7898305654525757, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/467a925c3cce13161bb798fcea482a4bcb118ecd1dbb0b9e73f2c3a47e4e588b.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "46bd5c651c933122ad3a646371ed54660686076971e95276bf5fea97c663291f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8c9a9a21284c5687bce616eb10e201c9a53524874518bb419ba1216be9da3001", "generation_seconds_batch": 23.654634250095114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 6.41180419921875, "genuineness": 4.469690799713135, "identity": 0.3249395489692688, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/46bd5c651c933122ad3a646371ed54660686076971e95276bf5fea97c663291f.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "46c6f8405cffa3343fd7ab98a25e465aed7d76138ae03d28e6c65b1f67bfc41f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "dd1b086e81c9f091a79368236c3f11d2f3a10eb9f7e4c4dd7b9ed5f89ce39dab", "generation_seconds_batch": 24.883227605954744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 5.974957466125488, "genuineness": 4.319066047668457, "identity": 0.6966574192047119, "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_WARM_reg_high/46c6f8405cffa3343fd7ab98a25e465aed7d76138ae03d28e6c65b1f67bfc41f.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "470bb6f1ba72befdd84594a0095fa53b5237e8e7c4e6d8f923dbcecd427ae385", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "bc4ceeb69b0f8caf8f69db6f814c0471c54a59de4933ad4e867c9925caa2aef4", "generation_seconds_batch": 24.605219756020233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.292664527893066, "genuineness": 4.143336772918701, "identity": 0.2487548291683197, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/470bb6f1ba72befdd84594a0095fa53b5237e8e7c4e6d8f923dbcecd427ae385.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "075d47579b073aa8e06cf74e0d6479885e14087b1c8d0e468429765659c658ab", "generation_seconds_batch": 16.95845624397043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 5.990842819213867, "genuineness": 4.18099308013916, "identity": 0.5312893986701965, "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_WARM_reg_high/4755d5aa45cb7b8b78649ff9d73c9ba7d56b0c73347a504b1a7d63ee14466874.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "47a1a3e83c96c1ae95f82b9496262fd26ff3b6cf7ba082701fbf9bad60270e6b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e51e748106928e6fe66bc379488de7b088c060e9b64cf1c86bac0b8d8d09e257", "generation_seconds_batch": 27.07980597706046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.372574329376221, "target_z": 1.9531249176599295, "ce": 6.284721851348877, "genuineness": 4.3127899169921875, "identity": 0.32523414492607117, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/47a1a3e83c96c1ae95f82b9496262fd26ff3b6cf7ba082701fbf9bad60270e6b.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very warm vocal delivery. 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": "8b8210e057287cc6eb7d61e2b9f82e29c1d5b845b9f95d32ccc4a0f78898cf68", "generation_seconds_batch": 15.925009939004667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.181467056274414, "genuineness": 4.607763767242432, "identity": 0.7259473204612732, "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_WARM_reg_high/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "4823562e39a3a3af981c43756ffa786ba2b0a9a0843eeab3893da13f96db982f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "37492537181ae131dc7d80614ceafdedfdec9b9780bd77a9a9ebc9c1a3e72fab", "generation_seconds_batch": 16.950606797006913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 5.569881439208984, "genuineness": 4.482243061065674, "identity": 0.7382822036743164, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4823562e39a3a3af981c43756ffa786ba2b0a9a0843eeab3893da13f96db982f.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "489f51f8689630cb1f1b81ae1333da2b7ed9d3961d28b8ca7a1e4d641eb4784f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "60839c2fdf181d6e0a9b8465b64559cd9a4539883dabb1d7315150432d6e0839", "generation_seconds_batch": 19.014681337983347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3015940189361572, "target_z": 0.769531328223067, "ce": 6.038498878479004, "genuineness": 4.432034492492676, "identity": 0.22552050650119781, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/489f51f8689630cb1f1b81ae1333da2b7ed9d3961d28b8ca7a1e4d641eb4784f.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "48b368d029cac6cbce02ffc864cbe21e9dd807bb7c9d1258afdc001c64df6026", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "28d202826e84b8a35ad1b17cc88f517eb058c4d955d36c5ce557aa9ddd109c5c", "generation_seconds_batch": 15.555132296984084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0683109760284424, "target_z": 0.5117186388409047, "ce": 5.919358730316162, "genuineness": 4.13078498840332, "identity": 0.7159470319747925, "hypothesis": "De deur 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_WARM_reg_high/48b368d029cac6cbce02ffc864cbe21e9dd807bb7c9d1258afdc001c64df6026.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "48c7748ed4e948344044f14e3ce01dd09a43220cf5b4236b941b34c79bf1a586", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ebe72949a450e8ad4e1b3c625865fdb89d3c48ee728f8d6d33752e9628f95b67", "generation_seconds_batch": 19.019884507986717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 6.332377910614014, "genuineness": 4.036643981933594, "identity": 0.22608399391174316, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/48c7748ed4e948344044f14e3ce01dd09a43220cf5b4236b941b34c79bf1a586.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "4945ddfb57404fed37ce5183019b9699a25521ba6f46130abbe5eaf11c4760a6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4900196e3afe10ce6cd95ecce51a3f0d4f35dff1269043097266e3b63a9df792", "generation_seconds_batch": 23.497221046010964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 6.022613525390625, "genuineness": 4.193545341491699, "identity": 0.7635841369628906, "hypothesis": "De deur 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_WARM_reg_high/4945ddfb57404fed37ce5183019b9699a25521ba6f46130abbe5eaf11c4760a6.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(very warm vocal delivery. 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": "a5041c08229c50f29ec48f19755cdf1e97cec806eb7dc083adcdec731500a3e8", "generation_seconds_batch": 19.497521262965165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.221180438995361, "genuineness": 4.231201171875, "identity": 0.7136843204498291, "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_WARM_reg_high/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "49ee287bdfc77e9400309fbfb3f7f1d4314480a17074553f177edcbf5fdb4e79", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "67f3e81b2e807eb6b437eb205477019fbff940ae26da45eee8a07e9c63951d1b", "generation_seconds_batch": 22.97818377194926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.602034091949463, "target_z": 1.1015625905740776, "ce": 6.109982967376709, "genuineness": 3.7918784618377686, "identity": 0.6231749057769775, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/49ee287bdfc77e9400309fbfb3f7f1d4314480a17074553f177edcbf5fdb4e79.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2932cf0877be29a9c691697c7f25ef1e847c76f85031d5204774fdf7b1acf46a", "generation_seconds_batch": 23.527283388073556, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.245328903198242, "target_z": 1.8124996706397178, "ce": 6.006728172302246, "genuineness": 4.300237655639648, "identity": 0.605716347694397, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lit le bien, and verifie les deux derniers chiffres. Appelle moi quand tu aurais a moment.", "wer": 0.22727272727272727, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4a3662be42b79ac4ffebc53f0daa4a842dd2ab3251d111baf292566ce37e4051.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "4a5619da39c4887041b17949373df8dc02f5fbb2e57226d2feb8fb9eac740cd5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ce71fe0ec9c5533188c8026aecdf54b911da5359f4f3754b0c291eb1c0bf028f", "generation_seconds_batch": 17.653588859946467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.149696350097656, "genuineness": 4.168440818786621, "identity": 0.7982298135757446, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4a5619da39c4887041b17949373df8dc02f5fbb2e57226d2feb8fb9eac740cd5.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "07f26605ab8c6ad75b98c8a4db475a340e85ae911bdadc6c9bd598c8fd78b40e", "generation_seconds_batch": 18.088549594045617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.284721851348877, "genuineness": 4.15588903427124, "identity": 0.6773407459259033, "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_WARM_reg_high/4a5a045606dbcefbdb0650d4ab782a44574879ee88f99ffc19ee964de752e487.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "4a84cea67dde86283e3a679a6a8b862be78df96c405c2fafd6a29a592661e0e9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "faea87e126ee8b3dd096da92e565ed62aeed5c435fec8a8c994bfcf81d3b9d07", "generation_seconds_batch": 17.37557801499497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.197352409362793, "genuineness": 4.174716949462891, "identity": 0.3184576630592346, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4a84cea67dde86283e3a679a6a8b862be78df96c405c2fafd6a29a592661e0e9.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a6fb654c13e1d08b893e1d309f8a82af58b6cd51c3a0aba9bd9b97bb05923a5f", "generation_seconds_batch": 21.360847497940995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.419747352600098, "genuineness": 4.168440818786621, "identity": 0.20833729207515717, "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_WARM_reg_high/4a8ce1449e79a578ac322c7a16f13089c6b9f134d74050cee2c8dedb8285a6ea.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "4ac0809f1f291fbfb8ccd3efcc0651bd30081a5e0766759a568257b59d14a51c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "fb35c813e09ac69e5f959a3396863fd28291ab1d3636fb4218a94a758f748746", "generation_seconds_batch": 21.559644137043506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.292664527893066, "genuineness": 4.432034492492676, "identity": 0.6094414591789246, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4ac0809f1f291fbfb8ccd3efcc0651bd30081a5e0766759a568257b59d14a51c.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "4ac42ff5b212f974d56719a8f67fc40f499d66902698b0898f6cf9bf33483712", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "67abfb97db821030f0b72e94ee0bd2eadc5f4e5b18b3b09abb4982c48769b54c", "generation_seconds_batch": 24.813943092012778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8706626892089844, "target_z": 1.3984374094259224, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.5773389339447021, "hypothesis": "De deur 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_WARM_reg_high/4ac42ff5b212f974d56719a8f67fc40f499d66902698b0898f6cf9bf33483712.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "4af2973b903c048edc836a64eec84ebbb0d0691e030a7dfff65a2d68dba9509f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5786291fc75b93667c656009a4690d0de448f0a5d4964333947e359a22fa6933", "generation_seconds_batch": 22.45012775505893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1849524974823, "target_z": 0.6406249835319859, "ce": 6.117925643920898, "genuineness": 4.118232727050781, "identity": 0.6996992826461792, "hypothesis": "De deur 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_WARM_reg_high/4af2973b903c048edc836a64eec84ebbb0d0691e030a7dfff65a2d68dba9509f.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "4b12799736b85efff653f26ecd0c9c4a2e565f591eda5b7a8e7f17c773b8ab68", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f972adb968cee0bc02e7ded87ac1839952440c876c541b79d289a5fe2f31a19d", "generation_seconds_batch": 22.45012775505893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4959964752197266, "target_z": 0.984375148212127, "ce": 6.09409761428833, "genuineness": 4.080576419830322, "identity": 0.6990811824798584, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4b12799736b85efff653f26ecd0c9c4a2e565f591eda5b7a8e7f17c773b8ab68.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "4b474a9aeda7a5e3ae2948955b6d9237e916888f370a22a921a228f5c803596e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4dcff25966543068fd9b9f4022fa1e30cc75403a5cdd08fb1f36a039d097f6a5", "generation_seconds_batch": 23.959343201946467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4182353019714355, "target_z": 0.8984374094259224, "ce": 5.903473377227783, "genuineness": 4.074300289154053, "identity": 0.6440944671630859, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4b474a9aeda7a5e3ae2948955b6d9237e916888f370a22a921a228f5c803596e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "74fa5fb29495694ffc9c1bb9bd1bd2d8bfa46dfee457aa188e5d9d4b4d08d09f", "generation_seconds_batch": 17.73100562498439, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.109982967376709, "genuineness": 4.457139015197754, "identity": 0.8284330368041992, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b2517f604ff7f8ad1d6863504a9d2964a0033c3ec30be6320131ea201fc0adc4", "generation_seconds_batch": 16.08214486797806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729279041290283, "target_z": 1.2421873106178378, "ce": 6.109982967376709, "genuineness": 4.3127899169921875, "identity": 0.7771687507629395, "hypothesis": "Te escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4bd4e4f4cca2f226ea9347ce057573761b1602483f366c5b0e3dd096a19a9d0d.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "4c5605a707916120f8b7b0fe085a832e5ef79e8b40fe76e88c1c16899d0a538b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "ab9272d6eec76f2db28b4775f14dfceba2ef8f65ce86ff199b8042150d37a92f", "generation_seconds_batch": 24.35213533404749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.488927125930786, "target_z": 0.9765624588299647, "ce": 6.2767791748046875, "genuineness": 4.080576419830322, "identity": 0.35892385244369507, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4c5605a707916120f8b7b0fe085a832e5ef79e8b40fe76e88c1c16899d0a538b.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "4ca0d640caa0edb190973ef296f15adec29792493b6659043cd21ec0828a99d6", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "42316de5c62e7f15bcc23f5a6456fdaae740ac157ffc0914298a2fd1e5ededc5", "generation_seconds_batch": 17.562399860005826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.4346477687358856, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4ca0d640caa0edb190973ef296f15adec29792493b6659043cd21ec0828a99d6.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "4cc88c3dba295e91bddbf61541b3a0f6281829767715718db1abd8124a17535c", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "39997d86ef3342e2343ab045fbe354a72261fdbf68aa43050ecc0991153b0736", "generation_seconds_batch": 17.110177177004516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4959964752197266, "target_z": 0.984375148212127, "ce": 6.109982967376709, "genuineness": 4.168440818786621, "identity": 0.6726219654083252, "hypothesis": "Ik 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_WARM_reg_high/4cc88c3dba295e91bddbf61541b3a0f6281829767715718db1abd8124a17535c.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "4cf521ff216169c15a318a5f63e7dafb395e4d28d6cdcb0d0e6f89e743533685", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a017245e8338fd4fdcd2a9bcfdf38cb74bd7f79480a0599c33a74fe61015e57b", "generation_seconds_batch": 21.997723033069633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8706626892089844, "target_z": 1.3984374094259224, "ce": 6.24500846862793, "genuineness": 4.337893962860107, "identity": 0.8348112106323242, "hypothesis": "La porte est encore encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4cf521ff216169c15a318a5f63e7dafb395e4d28d6cdcb0d0e6f89e743533685.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "4d0742f709d0272a745a35d6b236413e7b781072c292c6c1fe7b546f1cbc9307", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "8125ee7ee74b424166a2037723f3fb0f3f6c12964f690aa7a845a6e2a7a80147", "generation_seconds_batch": 26.145704962080345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 5.287916660308838, "genuineness": 4.381826400756836, "identity": 0.6919031143188477, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble.", "wer": 0.15789473684210525, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4d0742f709d0272a745a35d6b236413e7b781072c292c6c1fe7b546f1cbc9307.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "4d1dd8b7bb0b577df5f33a9493f78e9038c4625ef56093433a71c4211d1ff1b5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d97c95850bf4383cc1a1308cdc9c6033adba3501c862d7515c086f361960eb1e", "generation_seconds_batch": 22.63324682007078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.308549880981445, "genuineness": 4.344170093536377, "identity": 0.40818217396736145, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4d1dd8b7bb0b577df5f33a9493f78e9038c4625ef56093433a71c4211d1ff1b5.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "4d3e4e196f1531e37fc985ec3489b252d017aee54d343748a4ce847aae5cb959", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a1f92ff60325fc0c07b8c332d6a1c682879f7e221fa5ad57af7f931d9f61d89d", "generation_seconds_batch": 22.961900369962677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.054384231567383, "genuineness": 3.8546390533447266, "identity": 0.5216702222824097, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4d3e4e196f1531e37fc985ec3489b252d017aee54d343748a4ce847aae5cb959.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "4dad0c136898ed6b3a93a1bcf6408a9ceed44453afaae32e646f327b441efe40", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9ab32ac513ca5a8a27c783c1bbe2f6ed7132d9bc8813eb18948d0bd4e2ce867a", "generation_seconds_batch": 17.507038222043775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.427689552307129, "genuineness": 4.21237325668335, "identity": 0.5547789931297302, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4dad0c136898ed6b3a93a1bcf6408a9ceed44453afaae32e646f327b441efe40.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "4de46e28715381b3df5943e96d252f37e10fd70ccd11a0dcbf4f1d1bf22c9829", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "869f674bec00e69331dd27095244c9a00f207a154db0d9fdb5b3f8b9feb71496", "generation_seconds_batch": 16.0184059770545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471254348754883, "target_z": 0.9570313940951235, "ce": 6.117925643920898, "genuineness": 4.362998008728027, "identity": 0.3614250421524048, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4de46e28715381b3df5943e96d252f37e10fd70ccd11a0dcbf4f1d1bf22c9829.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "4de60d1b68a940d4066a8832cc4e0b5d18b20f9226353f8bc91d32872db60305", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "209a28fed8929de8fcb349a18bb370c81842b129799c2cfce993d1c669319b98", "generation_seconds_batch": 17.496656198054552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.165581703186035, "genuineness": 4.250029563903809, "identity": 0.4062819480895996, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4de60d1b68a940d4066a8832cc4e0b5d18b20f9226353f8bc91d32872db60305.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "532c0fc34e81b86226df9ae587cd92f5cd73d103e540edb2ec696c6cca8016c3", "generation_seconds_batch": 20.993672338081524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.316492557525635, "genuineness": 4.6203155517578125, "identity": 0.7836147546768188, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cb09a40af9bafc957e6df13638b4bb566ab143656a48d193a96abe869676f57d", "generation_seconds_batch": 23.46334828599356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3546128273010254, "target_z": 0.8281250494040423, "ce": 5.780362129211426, "genuineness": 4.143336772918701, "identity": 0.6251468658447266, "hypothesis": "El était ici près de la grande bibliothèque.", "wer": 0.6666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4e2e2a46046ff6895a40661ef1e306eb9ed80a162ec77ab345b211ddfd857b06.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "4eb3abc3475d3bb3ef5b217d6af8feb8609a067893fa5eecd8c13740c754633b", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "487416bd629289d87b07502b35ef849b091c725dbb7f53ba86068f4faa172456", "generation_seconds_batch": 17.653588859946467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4323737621307373, "target_z": 0.9140625247020212, "ce": 6.181467056274414, "genuineness": 4.268857479095459, "identity": 0.8178889751434326, "hypothesis": "Ik 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_WARM_reg_high/4eb3abc3475d3bb3ef5b217d6af8feb8609a067893fa5eecd8c13740c754633b.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "4ed0ec8f98557e52c4abc31537de8a8da35e512787d7aef4a16b0fa799f89967", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6c7884a72c4e094f795bf4204eb9ceda6f5cfd5f703ac3e1b3e6df0374b755dd", "generation_seconds_batch": 17.88476680603344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 5.9431867599487305, "genuineness": 3.9676074981689453, "identity": 0.5041681528091431, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4ed0ec8f98557e52c4abc31537de8a8da35e512787d7aef4a16b0fa799f89967.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "4ef907c38b722cb65d2a0967c51a107295317a537488cdcfc09dd67f4cc8b30e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b24558810564dc50ff74e1a0b47fc98f3c673f6d38f93684e799e43c4d073ea2", "generation_seconds_batch": 28.754313410026953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 5.637393951416016, "genuineness": 4.293961524963379, "identity": 0.7006661891937256, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle conçoit.", "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_WARM_reg_high/4ef907c38b722cb65d2a0967c51a107295317a537488cdcfc09dd67f4cc8b30e.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "4f6967d61594031584d359ae567caa4dc708dd29cea90e330bd55bb8a8253c2d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3320cb613357f819c0dc113cd8840d8c8c93f81d47e14996e52e4ea8bd3863b8", "generation_seconds_batch": 16.901269335998222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.884801149368286, "target_z": 1.414062524702021, "ce": 6.41180419921875, "genuineness": 4.281409740447998, "identity": 0.3088147044181824, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4f6967d61594031584d359ae567caa4dc708dd29cea90e330bd55bb8a8253c2d.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "4f7208c7b927345410e3aead8097f6906e65d93efd95ae4cfbb76e6a4f2278b2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "439e6d57d05985bb5e234b217144d89eda4fddbac8c0e85677bcbb215d01db32", "generation_seconds_batch": 24.101121773943305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8141093254089355, "target_z": 1.335937475297979, "ce": 6.316492557525635, "genuineness": 3.9048471450805664, "identity": 0.5665866136550903, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4f7208c7b927345410e3aead8097f6906e65d93efd95ae4cfbb76e6a4f2278b2.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "4f8f5434c55e81ad110817883eb79af431c50c54af228543cdf971bdb03826dd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "37e7239b6cbb68764054c0900fb3d7b7a2a13365d55edb6dd5c5e7ed9a678ff4", "generation_seconds_batch": 23.864452486042865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.499819755554199, "target_z": 2.093750164680141, "ce": 5.839932441711426, "genuineness": 3.9048471450805664, "identity": 0.5121197700500488, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuánto ha estado.", "wer": 0.2, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4f8f5434c55e81ad110817883eb79af431c50c54af228543cdf971bdb03826dd.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "4fa802a1c1e00523f5295f0a775e62cd42c67f2a507a86e7f97054e1925be443", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7b01cf9faa005289efda158b00232ef882016801e7fc588abe73dc36be3f70ae", "generation_seconds_batch": 23.966446034959517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.308549880981445, "genuineness": 4.494794845581055, "identity": 0.6758933663368225, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/4fa802a1c1e00523f5295f0a775e62cd42c67f2a507a86e7f97054e1925be443.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "501afea96e48decf1faeda80339511d1a4adebe39e7d3150a7cb7af7abfb54f6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "70491d7b8105f1a5c8caefbd1b1a2af2227605ba546f7eff6e06f0d914dbc566", "generation_seconds_batch": 15.634909820044413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.368751049041748, "target_z": 0.8437499011919153, "ce": 5.752562999725342, "genuineness": 4.268857479095459, "identity": 0.6623885631561279, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/501afea96e48decf1faeda80339511d1a4adebe39e7d3150a7cb7af7abfb54f6.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "506ae3d6997bfaf71ac00881f2d333fffd7f3de276d4204327b14c227201ef33", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7f680353bcb66e1148cf4069c9ceef3c53ab8ad5e6b259c949798d260c9eb0c7", "generation_seconds_batch": 24.874132041004486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.701002597808838, "target_z": 1.2109376070420916, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.7514026165008545, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen weer 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_WARM_reg_high/506ae3d6997bfaf71ac00881f2d333fffd7f3de276d4204327b14c227201ef33.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "5117141fb416ba3d46d25ce1c6fcce0e7c5989ff9d7ba32acd5e9812f404ee91", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "1e7366fa6c8f2e4d042669c0db9dc295854a0cfde38ddc25f9227417e7990e10", "generation_seconds_batch": 22.097469568019733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.485681533813477, "target_z": 2.078125312892268, "ce": 6.038498878479004, "genuineness": 3.9676074981689453, "identity": 0.746266782283783, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5117141fb416ba3d46d25ce1c6fcce0e7c5989ff9d7ba32acd5e9812f404ee91.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "5290d0835a97a901c6013bf5aad99257eb1ed9db8f24313b82732996a591a3f4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "39368c5271dbbaaed3cbe457eaac9b52a4544743a8c6f74dcc77eb2b7c5ee0f3", "generation_seconds_batch": 25.807430564891547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.308549880981445, "genuineness": 4.055472373962402, "identity": 0.4713841676712036, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5290d0835a97a901c6013bf5aad99257eb1ed9db8f24313b82732996a591a3f4.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "52b56e7d58ad26b07d6760612ae0586804a414e82da7ee3877ed58296807ac6e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "265c0553b7c55029f60d95f5d6a5e250a62159f578a44a225b5dadb91a3bd950", "generation_seconds_batch": 17.274533401010558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.062326908111572, "genuineness": 4.137060642242432, "identity": 0.5835349559783936, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/52b56e7d58ad26b07d6760612ae0586804a414e82da7ee3877ed58296807ac6e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "52d245f5231008996033df3d01fb40d3a140012163e6de889057715d1992b184", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fd706992d15bc86aa7b7ee01bca3c93bdc465b7a117f718015b5ecf3c3e44cc0", "generation_seconds_batch": 15.457227946957573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.213237762451172, "genuineness": 4.369274139404297, "identity": 0.7525132894515991, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/52d245f5231008996033df3d01fb40d3a140012163e6de889057715d1992b184.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "53489886b43001354549ae5b8240b3e03ca767e93600775660b4fabea0158364", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d96bf9b0f45ecd307eb93bc0502c4e950582c0864714622266cf99b4de287e77", "generation_seconds_batch": 23.449130148044787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.09409761428833, "genuineness": 4.3881025314331055, "identity": 0.7049139738082886, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/53489886b43001354549ae5b8240b3e03ca767e93600775660b4fabea0158364.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "534f11415c45d9b900600a67be20775f6bc9de1eb3ce62a6451b382acb69d3ac", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e6ae6383919ce33a4774d87cd56b5296002ea296547611767bd8e01700906ff3", "generation_seconds_batch": 23.242546002031304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.347543716430664, "target_z": 0.8203126235101058, "ce": 6.09409761428833, "genuineness": 4.281409740447998, "identity": 0.7602081298828125, "hypothesis": "De deur 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_WARM_reg_high/534f11415c45d9b900600a67be20775f6bc9de1eb3ce62a6451b382acb69d3ac.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "535448ac2ced832ffbd1d040a0b98ddce9cda34856b8905b3387182f08407743", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "35f8049a0c13f62a467186a74525320b44163349a217f7bcdfa1746d9f73ff6a", "generation_seconds_batch": 23.026826991001144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 5.8875885009765625, "genuineness": 4.143336772918701, "identity": 0.3780316412448883, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/535448ac2ced832ffbd1d040a0b98ddce9cda34856b8905b3387182f08407743.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7633375dc1a15160024ed3b538fef51a9e2244c54867cc54ab3031e31d6a5d6c", "generation_seconds_batch": 17.422137067071162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.523001670837402, "genuineness": 4.356721878051758, "identity": 0.2337794005870819, "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_WARM_reg_high/535ad7617f57062c7d0ed4cb4d67f13defa5dfb041ec4e7fd8ff6ec798ae8444.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "537215c506065fa49c428775986bdecddd61e9cc3f6f83e7df9a3543c99740af", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "a74a480e22c13008b9a0f4e0d9b664339f56214af1ebd39b9c204a49ff9880ac", "generation_seconds_batch": 24.714191720006056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.292664527893066, "genuineness": 4.017816066741943, "identity": 0.2707422077655792, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/537215c506065fa49c428775986bdecddd61e9cc3f6f83e7df9a3543c99740af.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "5387525ea818f83a41764f4771e1b575f521f027c54e0c582c46e0889da76041", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "830d8dded02bc2b13466bf7eaafb24d99ddb6d143406c1e286bfc363ab4701e9", "generation_seconds_batch": 16.458869230002165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.41180419921875, "genuineness": 4.281409740447998, "identity": 0.6400593519210815, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5387525ea818f83a41764f4771e1b575f521f027c54e0c582c46e0889da76041.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "53b962b78582d4d5db096d2cf1cd79dc06794f22afe08ed4a7bc346c931e4f59", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "77e49a1c21c40b067913c6a1dee5a93c2d030b9e9b1fe7ed3db3f5c431360e2c", "generation_seconds_batch": 15.858937766985036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.149696350097656, "genuineness": 4.005263805389404, "identity": 0.6628358364105225, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/53b962b78582d4d5db096d2cf1cd79dc06794f22afe08ed4a7bc346c931e4f59.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "53c1559cd93b176d074051ffcb29888b18eacec618975389af8f78842eb9be7f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "5748fc474e22aa934659460b2aa67323087031f3f6bc8915532338e757fd15ee", "generation_seconds_batch": 16.229844363057055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.24500846862793, "genuineness": 3.8671910762786865, "identity": 0.3771335780620575, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/53c1559cd93b176d074051ffcb29888b18eacec618975389af8f78842eb9be7f.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "54212023ee42baa86f9568fcc3db4e0143f319ccea0cc9f1abf01fa24053189f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b8e9b8a94f4aa3046be0e62d15dc3173e16b36f6ab23c682c5045a9016934ec7", "generation_seconds_batch": 22.97818377194926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.260893821716309, "genuineness": 4.243753433227539, "identity": 0.6904696226119995, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/54212023ee42baa86f9568fcc3db4e0143f319ccea0cc9f1abf01fa24053189f.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(very warm vocal delivery. 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": "6f254406f6a9918c4d5483ee08e1bd096f0f4bc604555e65b60eb0ee2c815b06", "generation_seconds_batch": 17.999235155060887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2202982902526855, "target_z": 0.6796873764898942, "ce": 6.229123115539551, "genuineness": 3.672633647918701, "identity": 0.5843441486358643, "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_WARM_reg_high/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "5473042b59138eb25c14bfde3bbadc43d915a1ffcf769a0c5f92c5607b11bc6b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5197b8e9f42c13fd345e417bdaf66d7136a8391a838bff05101a919d2174e4da", "generation_seconds_batch": 23.278218221035786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 5.450741291046143, "genuineness": 3.986435890197754, "identity": 0.688020646572113, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5473042b59138eb25c14bfde3bbadc43d915a1ffcf769a0c5f92c5607b11bc6b.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f7b8e9298cb76b5cebb9a873db9d55ed9a9ef534a6e483dcde2f06203d87f635", "generation_seconds_batch": 20.40659520495683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.195845127105713, "target_z": 1.7578126893821622, "ce": 6.149696350097656, "genuineness": 4.413206577301025, "identity": 0.4457378685474396, "hypothesis": "Don't message arrived after my départ. Je l'ai relue, puis j'ai rangé mon telephone. Dis-moi quand nous pourrons parler.", "wer": 0.3333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5499c8cf65ff410baa0e4f85a52969187e6726c017dec4ed958c2de2eec41ea4.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "54a39444bd4fa1ca631ba28be027f1a3d7c5dbccb564b89cf4cd28e80782953e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "202273e3019ced09a519bcbe82510516e1a130b1bb730ef421e0209a4f54283b", "generation_seconds_batch": 24.363136643078178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.300607204437256, "genuineness": 4.21237325668335, "identity": 0.24080397188663483, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/54a39444bd4fa1ca631ba28be027f1a3d7c5dbccb564b89cf4cd28e80782953e.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "54ec6435be38260d8826a5cf68efc4ce41caa52ae6e044dcbc6df7998709cfc8", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "84edf95745a9df031d8e8c98d48328ab63c6279b976d712d1cb08aabb827ff34", "generation_seconds_batch": 16.70772501709871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.173524379730225, "genuineness": 4.243753433227539, "identity": 0.4774817228317261, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/54ec6435be38260d8826a5cf68efc4ce41caa52ae6e044dcbc6df7998709cfc8.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "458f24e7ac08b61fe0051642fbba32eac95a29bc3a648757d61e9b11994cada5", "generation_seconds_batch": 18.942736728000455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089807510375977, "target_z": 1.6406252470202116, "ce": 5.8875885009765625, "genuineness": 4.369274139404297, "identity": 0.6348017454147339, "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": 0.120000000000001, "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_WARM_reg_high/5553acb23b9490248eef8c5d8752200f1e3dde385defa35e5f4ce3a4bc96e311.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "55628c6ba31341145e4b6bc14cdcadea09c2a75e951f64670e8c141b34143c53", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "70db65f28099a1fe1d93d1cefc163854fa0ba9359b9929e31e65cbdbe91f4834", "generation_seconds_batch": 23.931804548949003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.372091293334961, "genuineness": 4.243753433227539, "identity": 0.5580092072486877, "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_WARM_reg_high/55628c6ba31341145e4b6bc14cdcadea09c2a75e951f64670e8c141b34143c53.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "55b612694d91a477116c466e789fcd78804e9d7e5e86cfc8dee1bcd3f8895a80", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e1b6bbd144eb3b3619f906905241b80b8f333edb2b00eb6390ee46c3fd45500d", "generation_seconds_batch": 16.950606797006913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2099833488464355, "target_z": 1.7734375411700354, "ce": 6.284721851348877, "genuineness": 4.770940780639648, "identity": 0.857277512550354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/55b612694d91a477116c466e789fcd78804e9d7e5e86cfc8dee1bcd3f8895a80.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "55bab9366dca661b73cd6e8d67af370adfe4bb4f5bbe830b1577c4388d3b37e6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "315c36b06675d487f92da96104a014c4dc824a124626ea6d9e36f8d0f28d9eb2", "generation_seconds_batch": 24.219300976023078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.141753673553467, "genuineness": 4.005263805389404, "identity": 0.6664465665817261, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/55bab9366dca661b73cd6e8d67af370adfe4bb4f5bbe830b1577c4388d3b37e6.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "55d25334b738d633a878b2645495663057f2abd38a5c42abeb34ca650d0d1a69", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "bf461f308068dfd04cc98d8402bdaea32f01bf5ec84305c727fdaf754ef3aae0", "generation_seconds_batch": 18.43187206494622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.149696350097656, "genuineness": 4.168440818786621, "identity": 0.7209838628768921, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/55d25334b738d633a878b2645495663057f2abd38a5c42abeb34ca650d0d1a69.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "56118efd09f5be484b4a98192d49c459962d1e8a255b309e031fb4ff9a2663ae", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "77696963450d52baea5913cbe0ed76016c8a4531ca89490742b0d4935c3461f0", "generation_seconds_batch": 23.95174243801739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.205295085906982, "genuineness": 4.369274139404297, "identity": 0.4456177353858948, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/56118efd09f5be484b4a98192d49c459962d1e8a255b309e031fb4ff9a2663ae.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "5613bf557cf58c1ec3a80e6d3934223e02782f82f5610944505314a7c1afae23", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f2d0c20aacfe6108be39a563707d9537aafbaf884e5616e9312174d547c589b4", "generation_seconds_batch": 25.920054264948703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 6.197352409362793, "genuineness": 4.507347106933594, "identity": 0.7882022261619568, "hypothesis": "La porte 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_WARM_reg_high/5613bf557cf58c1ec3a80e6d3934223e02782f82f5610944505314a7c1afae23.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "564251bcc2742a1f51c80b6f73b6f22406ab72e36019e581330dfb03c18bf411", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "321f7ed1be8f8bf3c4f1f6071279566bd40d5640005af3b485a1daf8e95e46fe", "generation_seconds_batch": 15.446155859041028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 5.95112943649292, "genuineness": 4.3504462242126465, "identity": 0.8254834413528442, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je kuurt.", "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_WARM_reg_high/564251bcc2742a1f51c80b6f73b6f22406ab72e36019e581330dfb03c18bf411.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2964ab75f39cc0dd89853dcb834bf643a54157a05cb175153dd1c795159c74f9", "generation_seconds_batch": 17.336190436035395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.109982967376709, "genuineness": 4.375550270080566, "identity": 0.7752130627632141, "hypothesis": "La carretera girerá a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/56760fe2d7110744aeed7c1a7f5f1cc0a2fa8902044b1f0b102dc897af9107f9.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "56aab34bfbdcc6e9cb3878eda9e9c156efea29b0b0dd27f3e53917bd9113b833", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "52989f348a0b431626286f5d9a9a69d71f339dea0441e686e265c0ad2f081490", "generation_seconds_batch": 17.930977203068323, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.109982967376709, "genuineness": 4.519899368286133, "identity": 0.6542749404907227, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/56aab34bfbdcc6e9cb3878eda9e9c156efea29b0b0dd27f3e53917bd9113b833.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "7749f1154d56eb5f0ed68d2edba89c8ca30ce9db4807098ba8999f53ccd8fe44", "generation_seconds_batch": 16.08214486797806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.5068377256393433, "hypothesis": "Escribe el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/56b1d87377627e245f4d9c4fa037ca1d7726c16ba5a2150921a2595b8ed3d1fc.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "573aaf2f99e0a4c7fbd3817b3249bfaf528eb97bdea88bbdf2ec3b8e23a952cd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7b833a20ca51e7c3ce2f7454d6ab06963069d2658f92289fbba75de21d88d7a9", "generation_seconds_batch": 22.63324682007078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.395919322967529, "genuineness": 4.406930446624756, "identity": 0.30305764079093933, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/573aaf2f99e0a4c7fbd3817b3249bfaf528eb97bdea88bbdf2ec3b8e23a952cd.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "57567f3e6936cb90032a6a3818ae38af81d66264518c93790f1a48791adea47d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "99c01b0014e5177e7d4db547168d5d089fc602544594a7f8a2223410b9eab202", "generation_seconds_batch": 17.110177177004516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.141753673553467, "genuineness": 4.3127899169921875, "identity": 0.6182624101638794, "hypothesis": "Ik 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_WARM_reg_high/57567f3e6936cb90032a6a3818ae38af81d66264518c93790f1a48791adea47d.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "5817e1a682da8610fdc452e33a48d5c596fe4b8cf89db3bd90984d9746e62ca0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7ae71a50cd40ffd01bb6b345f9aeb4f9a56a64589063c6792280519c94eaaa72", "generation_seconds_batch": 22.393970516044647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 5.990842819213867, "genuineness": 4.356721878051758, "identity": 0.7586771249771118, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5817e1a682da8610fdc452e33a48d5c596fe4b8cf89db3bd90984d9746e62ca0.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "583f23b2acfd5639ca6f28ae4dcf8d1f940e5f16c2224657382065e688f604fe", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "46562c05b9f6e6d089cc3fa7b4b4a3f1c9c27662bccaea78066c46bf7ca96f92", "generation_seconds_batch": 23.183197143953294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.41180419921875, "genuineness": 4.369274139404297, "identity": 0.38601547479629517, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/583f23b2acfd5639ca6f28ae4dcf8d1f940e5f16c2224657382065e688f604fe.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "584e188388b6a437416bf4f62f32eb41833194a689bcafcfc7d1fe8233e9d672", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "14c1cf9b6f889a103f7a0761d3bc5985561998224dbcec1d320fd5f54f264110", "generation_seconds_batch": 22.90902197198011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.372091293334961, "genuineness": 4.293961524963379, "identity": 0.6315470933914185, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/584e188388b6a437416bf4f62f32eb41833194a689bcafcfc7d1fe8233e9d672.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "58b1807ca1e6ac5c5569dfcb269459c30c80f286dc1c2eedfcb45a511c83a942", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3dd0157139b9cac6a127334eb361dd03decb12da687ab9e21ac45d84354b0841", "generation_seconds_batch": 28.555501414928585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 5.446769714355469, "genuineness": 4.406930446624756, "identity": 0.8168458938598633, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/58b1807ca1e6ac5c5569dfcb269459c30c80f286dc1c2eedfcb45a511c83a942.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "58d78513930fcecc5f4a5df273779070ae3b0a838afaa9809b61227fa281fd0c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1753b18733730105727b7017eb88637427b5b57b0a28021345de9637b1d44881", "generation_seconds_batch": 23.19145959895104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.3641486167907715, "genuineness": 4.356721878051758, "identity": 0.6523942947387695, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/58d78513930fcecc5f4a5df273779070ae3b0a838afaa9809b61227fa281fd0c.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4937eb5500a4db8796f30f56b5be1d75e84c11517caac560c90834eead0967b0", "generation_seconds_batch": 20.988031437969767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.217052459716797, "target_z": 1.7812499670639719, "ce": 6.062326908111572, "genuineness": 4.18099308013916, "identity": 0.5936422348022461, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un número.", "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_WARM_reg_high/590ba37ece575739b7705ce7a9590f5279730db961e59c0b5419dc7218f06a85.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "5976e2530c898a22375b97305206ccb8cd35a5a641218e801b7ca9188aa8bbf3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "dfa8c603e4442785d1cc1c8dbd9aa9ff9468d1fd95937d512b14ac0513b8e894", "generation_seconds_batch": 23.95174243801739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.221180438995361, "genuineness": 4.331617832183838, "identity": 0.6335806250572205, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5976e2530c898a22375b97305206ccb8cd35a5a641218e801b7ca9188aa8bbf3.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "5a55bfcb0387c7a5384554900fb527c04cf5bec129f845dde88d3b864ce2c3fe", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a471fa54308aa61269c72f24c4e8cda2ecfc0cdf3eee1f4750d823214192fbc9", "generation_seconds_batch": 22.436538369976915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.0146708488464355, "genuineness": 4.306513786315918, "identity": 0.6409310102462769, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5a55bfcb0387c7a5384554900fb527c04cf5bec129f845dde88d3b864ce2c3fe.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fee7125aa634ae2ada8d574752f20158a435cfdd03091eb2ba1c112c761ba37f", "generation_seconds_batch": 23.325941888033412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2311906814575195, "target_z": 1.7968748188518449, "ce": 6.086154937744141, "genuineness": 4.15588903427124, "identity": 0.7008355855941772, "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_WARM_reg_high/5aca90c9d1ab71e1a6cfad83b0f6da9b0e750dc2df9e844c62be14a84b30334b.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "5b1f6a4461e00411b6c0fa8fb92cb1a120898d0c25474840e5a0f2977d82cadb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "315313c409d2741ce9e18e4e81d31ae05ef8d16a9a6124203660ee69d1395e1b", "generation_seconds_batch": 21.775823137024418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.308549880981445, "genuineness": 4.18726921081543, "identity": 0.3188022971153259, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5b1f6a4461e00411b6c0fa8fb92cb1a120898d0c25474840e5a0f2977d82cadb.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "5b8d1a636c7c9d73a73e8592af570c2d475dae0d6265a124c8b8d1db10df3bab", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8a24dbaff29f72b536cc0848bc7d4bb0e267ecda05c271888a94d0f60a45cb98", "generation_seconds_batch": 22.521420850069262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.483288764953613, "genuineness": 4.482243061065674, "identity": 0.4156392514705658, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5b8d1a636c7c9d73a73e8592af570c2d475dae0d6265a124c8b8d1db10df3bab.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "5bb3bc0e5c0bb094a9026b3e3c5e9dbe1f2475b85bce396dbffb82824db3ccfc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "929c54426e84071796d6a7d7d6da035efa0b3608e9b4f70317cd16f28bd81ba7", "generation_seconds_batch": 23.44711526297033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.701002597808838, "target_z": 1.2109376070420916, "ce": 6.197352409362793, "genuineness": 4.494794845581055, "identity": 0.6675915718078613, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5bb3bc0e5c0bb094a9026b3e3c5e9dbe1f2475b85bce396dbffb82824db3ccfc.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "5c17466cacff15d8d41723bab672767a4631d0a34969f823372349d46eda7d20", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "982d0fb6615803b3638fa7cc0ef4f68df79edae04ffc256f179821a4036fcda4", "generation_seconds_batch": 17.00631051498931, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4182353019714355, "target_z": 0.8984374094259224, "ce": 6.24500846862793, "genuineness": 4.162164688110352, "identity": 0.18269728124141693, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5c17466cacff15d8d41723bab672767a4631d0a34969f823372349d46eda7d20.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "5ccb530ab3a5b07073de6d72e727c27786923ee08a8a74c600f26e30f88ad019", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e7dd3d099521c9eaf4d6f85bc5e732a3a223814d082d3989c5173ee832c2ae62", "generation_seconds_batch": 17.23486279696226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.305128574371338, "target_z": 0.7734375411700353, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.7179771065711975, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5ccb530ab3a5b07073de6d72e727c27786923ee08a8a74c600f26e30f88ad019.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "5d9976eaceb0e96f5b41225d84eb6cfd0ba7e3035c82858f1ea04adb4a0bae8d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0422bee1a7954f4579f03ddd9750ebcd6a4d932f08420664a34fbfbf4965698c", "generation_seconds_batch": 22.17250768793747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 5.919358730316162, "genuineness": 4.595211505889893, "identity": 0.7564060688018799, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5d9976eaceb0e96f5b41225d84eb6cfd0ba7e3035c82858f1ea04adb4a0bae8d.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "5dad993934a9ae7ef203606dfacc5f9134ee69054998acc452c144d6204363a5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "11a47303c9ea818f72908252a0cb1701533d75248ea7c704357f8a2c0f9b3a63", "generation_seconds_batch": 24.43682372092735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.268836498260498, "genuineness": 4.18726921081543, "identity": 0.6230471134185791, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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_WARM_reg_high/5dad993934a9ae7ef203606dfacc5f9134ee69054998acc452c144d6204363a5.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "5dbd380faededdfb3ad3e83982e881963f767b22e988d7363ba14fe90431dcae", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "63112e32a41bdd724cd8a6773d5683b4333e0e5abafa8bb77288fa5a0d74fdd2", "generation_seconds_batch": 22.365867875050753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.35073018074035645, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5dbd380faededdfb3ad3e83982e881963f767b22e988d7363ba14fe90431dcae.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "5dfc4719d01c5dc7478036c877432e2a3f6d77b2934a9db8b3d1ceb8e445dc11", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "483da258236f98d28ed6f0531743d5ea0a81e408cea575aa770b29f550375d28", "generation_seconds_batch": 24.101121773943305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686864137649536, "target_z": 1.195312491765993, "ce": 6.427689552307129, "genuineness": 4.099404335021973, "identity": 0.43788212537765503, "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_WARM_reg_high/5dfc4719d01c5dc7478036c877432e2a3f6d77b2934a9db8b3d1ceb8e445dc11.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very warm vocal delivery. 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": "bc8268ae1acc494c552905cd84d58a3c9f49e90f03ce76eaee5164f66f687961", "generation_seconds_batch": 17.401499514002353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.332377910614014, "genuineness": 4.532451152801514, "identity": 0.8283527493476868, "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_WARM_reg_high/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "5e3887b7d686f037bda9e77b9f7f23aa689a1ed9f1bcba9a6b99533e8dfaf31b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e7a0b460bf9d1b53528690175146041af1579bf8eaedb7626da46b0acf4d28fe", "generation_seconds_batch": 23.376218984019943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 6.395919322967529, "genuineness": 3.948779582977295, "identity": 0.4665318727493286, "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_WARM_reg_high/5e3887b7d686f037bda9e77b9f7f23aa689a1ed9f1bcba9a6b99533e8dfaf31b.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "5e62e44213a25efd66be5ddd00c8fb4be5994f9105e2589d1354d28f2bb0f8b2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4692720ca621eb29d54f78f64c16d81052b1de6d45253d4a1aac5a94078ee8f3", "generation_seconds_batch": 18.428318747901358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 5.895530700683594, "genuineness": 4.381826400756836, "identity": 0.7065259218215942, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5e62e44213a25efd66be5ddd00c8fb4be5994f9105e2589d1354d28f2bb0f8b2.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "5eac389a9da7694c90068a20ccee2ef64e6076ae543aec4037cb54a0aac1de72", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8686c25bc33ba6c1eaef05f1f05d18ae9e06ebf6675cea2157859a8ddfbed4a5", "generation_seconds_batch": 24.399975895998068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269782543182373, "target_z": 0.7343748847239012, "ce": 6.260893821716309, "genuineness": 4.494794845581055, "identity": 0.636395275592804, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5eac389a9da7694c90068a20ccee2ef64e6076ae543aec4037cb54a0aac1de72.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "5f23a807d8d62d1bea05ca89c72d61eaebce1d58eec22cd975324fb14b2da093", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1f5b45bad4d3efc285b90c335f738c38778d0ba08e4b56a8c885d822c14b0cad", "generation_seconds_batch": 25.480022610980086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.785149335861206, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5f23a807d8d62d1bea05ca89c72d61eaebce1d58eec22cd975324fb14b2da093.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "5f5fe8c32642b0935443ad0781b977d2962a5c7c693b8689d87f7cd0de8de9fb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a0741d84168c89355a1f584edffb34db5d5ea8a73b46454b66fa26080468ca01", "generation_seconds_batch": 24.08157822699286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 6.459460258483887, "genuineness": 4.080576419830322, "identity": 0.5690337419509888, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/5f5fe8c32642b0935443ad0781b977d2962a5c7c693b8689d87f7cd0de8de9fb.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "60900b2004ce5044f07cf886324bd36a41f86a2bcebcb65707ec50f2b53b3fcd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a29fa8a02ec595fae68be3963cc1179d3f306424911395361235aba9beeb985c", "generation_seconds_batch": 16.684433295042254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.316492557525635, "genuineness": 4.331617832183838, "identity": 0.27481809258461, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/60900b2004ce5044f07cf886324bd36a41f86a2bcebcb65707ec50f2b53b3fcd.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "609394d7f3adcc6503dd87841325f81288b64859b523f68bab1b8990b652be92", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e3787e3189ac62262ba68fb6a3e8443e0a6cb522df82679a8f1f2ba49885d670", "generation_seconds_batch": 15.410423223976977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.292664527893066, "genuineness": 4.532451152801514, "identity": 0.692418098449707, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/609394d7f3adcc6503dd87841325f81288b64859b523f68bab1b8990b652be92.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "6148d71ed41fb9843a2328e0934e89472fa6fc27e13f9096c0678277f43be2e9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4b516dc186d4bd1b6e5b4685e22e686ecfa002ab735ecac2c34548fed1d0cf0a", "generation_seconds_batch": 24.051191041013226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 5.879645347595215, "genuineness": 4.6203155517578125, "identity": 0.7448580861091614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6148d71ed41fb9843a2328e0934e89472fa6fc27e13f9096c0678277f43be2e9.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "6173289e6264df1ae67583cf950d935c58ae3e86b6c85c30d9205e0325972db4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5ab7847c8421396cb29b57fd878cf123049774fd41dc252620930f406c035362", "generation_seconds_batch": 24.774293038994074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2627134323120117, "target_z": 0.7265624588299647, "ce": 6.308549880981445, "genuineness": 4.105680465698242, "identity": 0.6558808088302612, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6173289e6264df1ae67583cf950d935c58ae3e86b6c85c30d9205e0325972db4.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "617a5d693f30a9f3b59df8ab3f2fa444b46abced90b1249391249a5a2b4b20f9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0da85e05081b8ee17ac0f6c8540907884880434b236cee687f3fa611fb8eb497", "generation_seconds_batch": 23.88352804200258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 5.744620323181152, "genuineness": 4.306513786315918, "identity": 0.8555546998977661, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/617a5d693f30a9f3b59df8ab3f2fa444b46abced90b1249391249a5a2b4b20f9.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "62bc4339d9b08caefb7e6196235e58590003177819c5fbe01cc2f018283ba328", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "50c9601b1487c34fb4dcc4f289586db8c452cf89b3ca969a017ca51025915abe", "generation_seconds_batch": 21.59061174991075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446512222290039, "target_z": 0.92968763997812, "ce": 6.046441555023193, "genuineness": 4.319066047668457, "identity": 0.7500417232513428, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/62bc4339d9b08caefb7e6196235e58590003177819c5fbe01cc2f018283ba328.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "af03ce93d3e17586272d75ea6c231a5dbe2ba36105fb2e9feebc1c9d38c40926", "generation_seconds_batch": 17.187940155970864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 6.3641486167907715, "genuineness": 4.162164688110352, "identity": 0.7311985492706299, "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_WARM_reg_high/62bdf71acef6aebbddf456fc23d278dddb4223290bb8d1c954ed8d51b366035c.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "63d095756dd7203423135e498828b5589e3bd20431ff822f28c1e8a7e80e06e3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "025f30fe3655a7b7f0d2d56e2d9876f66d29cc632f20eb664568df23adb69751", "generation_seconds_batch": 15.963248020969331, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.109982967376709, "genuineness": 4.293961524963379, "identity": 0.6627987623214722, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/63d095756dd7203423135e498828b5589e3bd20431ff822f28c1e8a7e80e06e3.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "6458aac33519af03f34f80c533536846cc67a90f118f95f0ce6b830057e2b93a", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "518143e9b0516fad21f180231171266dd454b9cb8710a37e80ffed309405f2f4", "generation_seconds_batch": 17.254701825906523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.395919322967529, "genuineness": 4.494794845581055, "identity": 0.48058462142944336, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6458aac33519af03f34f80c533536846cc67a90f118f95f0ce6b830057e2b93a.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "648748dfdb739f098a3a33606891e10df370596fca7889da0b398c04ef12b636", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ed92080c5c1ab9bf3591b30a04217f00bbebf65c25270ccd33a240e4d0be52f3", "generation_seconds_batch": 24.481939404970035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.181467056274414, "genuineness": 4.068024158477783, "identity": 0.7402420043945312, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/648748dfdb739f098a3a33606891e10df370596fca7889da0b398c04ef12b636.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "64af4e38c78ef09f28ab9b858099833c6973d86b8d54f7787ca73c0a98e07ca9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9bf1f3b0f3c83d45db9cbd812a946740d1ccc9790340ff86e4f3ec7d4109c383", "generation_seconds_batch": 22.47985800402239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5066001415252686, "target_z": 0.9960937870530318, "ce": 6.316492557525635, "genuineness": 4.124508857727051, "identity": -0.004655160009860992, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/64af4e38c78ef09f28ab9b858099833c6973d86b8d54f7787ca73c0a98e07ca9.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "6507f197a150108af1b4cfdeda194ef0889dcefb725e6deec684043e8c0ea275", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0106ca1ef7678924f59c34bf6ac36f6eb69d4aab7bbb24e90bac7f1bca09984c", "generation_seconds_batch": 22.436538369976915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.301882743835449, "target_z": 1.875000131744113, "ce": 6.078212261199951, "genuineness": 4.406930446624756, "identity": 0.7038776874542236, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6507f197a150108af1b4cfdeda194ef0889dcefb725e6deec684043e8c0ea275.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "65116c8ae6d3c573b3506b55b6bd5c6fe6319a92bcc962d742fbb06c155ace4a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "49995baae8fd104a03564d07d5400e660cacbd956e6824b6109826019764d364", "generation_seconds_batch": 23.6397933709668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.173524379730225, "genuineness": 4.281409740447998, "identity": 0.5343344807624817, "hypothesis": "Die Tür 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_WARM_reg_high/65116c8ae6d3c573b3506b55b6bd5c6fe6319a92bcc962d742fbb06c155ace4a.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "6568fbdabdec94efa0294687cd9c680eb31f0cb4fb09c288ac58969ca912ff31", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b9fbce0916d140436c6d7602b9e2c2dee1f8a411aee58a6cae6d73f3d1a67ec", "generation_seconds_batch": 24.172303588013165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 6.260893821716309, "genuineness": 4.337893962860107, "identity": 0.7915737628936768, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6568fbdabdec94efa0294687cd9c680eb31f0cb4fb09c288ac58969ca912ff31.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "65a4cc160278ef45b1e4f0cf53e42f19dc5294e44147cf39e3e97f689635f313", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d6c95d77192525d02b04e36e36076c9b62ce72ed9e148c392f6c25d67b0f6b01", "generation_seconds_batch": 22.761452679056674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.556373119354248, "target_z": 2.156250098808085, "ce": 6.038498878479004, "genuineness": 4.300237655639648, "identity": 0.6552783250808716, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que se ha visto.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/65a4cc160278ef45b1e4f0cf53e42f19dc5294e44147cf39e3e97f689635f313.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "65a821b825b7a5c73d8997bed8526ce3c49b6d2dc0bbb45441acf8020af039d2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9f9a3e6a589b9144a9a602fff811c08607d693e2b6ed90e719faddf03b8a2c93", "generation_seconds_batch": 18.172895153984427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.51961350440979, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/65a821b825b7a5c73d8997bed8526ce3c49b6d2dc0bbb45441acf8020af039d2.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "67477be6047d16327679f9e96fc7a8a7d192c321219786161464064476f546da", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9d6b995624a8762693af8aa94787ded394107146b3230e08be79effa34fb6951", "generation_seconds_batch": 28.663059689919464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.403861999511719, "genuineness": 4.432034492492676, "identity": 0.4938756227493286, "hypothesis": "Ik 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_WARM_reg_high/67477be6047d16327679f9e96fc7a8a7d192c321219786161464064476f546da.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "67751a9bfec5c57ce8168017fd5d97df2c7335e26ae55e008990314cfba9647f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "40eac3a38cbe2203d38da7c7d3734797796e0376e54a784d4dc57895bdb76167", "generation_seconds_batch": 16.0184059770545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.340320587158203, "genuineness": 4.369274139404297, "identity": 0.5891671180725098, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/67751a9bfec5c57ce8168017fd5d97df2c7335e26ae55e008990314cfba9647f.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ac7ecbc38fd0a91e48874d88e93fd83236088ee47c1ce6b19a8d7f1d73fbc7a2", "generation_seconds_batch": 16.52397115190979, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.040323257446289, "target_z": 1.5859377387862046, "ce": 6.23706579208374, "genuineness": 4.507347106933594, "identity": 0.7681422233581543, "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_WARM_reg_high/6813725134bddb6381b1b71baa7eeedc17331e263a4cfb579546393557e1bf41.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "6844cb601429eb751843671587aef298b92c0d8eabe02bfe76a25aa4afa79ec1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6e27bd90590d418c6cd7919c5eef0898eb990c1d01c587e7f40dd1bffa9fc896", "generation_seconds_batch": 17.49666249193251, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8706626892089844, "target_z": 1.3984374094259224, "ce": 6.149696350097656, "genuineness": 4.030367851257324, "identity": 0.7207551002502441, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6844cb601429eb751843671587aef298b92c0d8eabe02bfe76a25aa4afa79ec1.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "685fb0c445f3f5c6a582705774644c8c7948b9a784f83f118a89725c0ebd4f0e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "666df318c1f7b30d53cdf5dd6317d2a73995e2cd5e12193d24568a7fc0e1e7bc", "generation_seconds_batch": 22.75128999003209, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.273605823516846, "target_z": 1.8437499011919154, "ce": 6.1020402908325195, "genuineness": 4.469690799713135, "identity": 0.6361088752746582, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que es.", "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_WARM_reg_high/685fb0c445f3f5c6a582705774644c8c7948b9a784f83f118a89725c0ebd4f0e.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "68aa726387728d707d1792e11399544c318f73fbffd10f328c4b3bf829650945", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c7c1c130e8fec8f407e6546d587155bac6a70cd2a11ab46763b5971faa9806f1", "generation_seconds_batch": 16.24828257295303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.273605823516846, "target_z": 1.8437499011919154, "ce": 6.133810997009277, "genuineness": 4.494794845581055, "identity": 0.6379004716873169, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/68aa726387728d707d1792e11399544c318f73fbffd10f328c4b3bf829650945.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "68ad4b230c4a9e65c78bac4e081d16e4fbaf6d5e570c067dd11cdbf42a1fcc90", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c5d4895923ebc01b62ce4d659d74cfc200098de5716208ecbd961e52cc8b4d75", "generation_seconds_batch": 22.855786213069223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.109982967376709, "genuineness": 4.20609712600708, "identity": 0.7783616781234741, "hypothesis": "De deur 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_WARM_reg_high/68ad4b230c4a9e65c78bac4e081d16e4fbaf6d5e570c067dd11cdbf42a1fcc90.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "49f3a6011bba359d1fb5afaea04c55276c939a0b56cf83c892fc9a6b657e6b30", "generation_seconds_batch": 16.52397115190979, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.884801149368286, "target_z": 1.414062524702021, "ce": 6.062326908111572, "genuineness": 4.331617832183838, "identity": 0.5642977356910706, "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_WARM_reg_high/68f38954792c32cc8b355a20d74dbbbc879fb756b285ed2ad192ea67b4b3adbe.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "698343180500a03f2496552c10165fda44bc9ebc0e9f8fd32a68347922ac39cf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "40449f8fb063df540333a965bf4ef65d8449f76ed7a3febf36ee924305222b07", "generation_seconds_batch": 24.939703290932812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743417739868164, "target_z": 1.2578126893821622, "ce": 6.268836498260498, "genuineness": 3.7165660858154297, "identity": 0.221711665391922, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/698343180500a03f2496552c10165fda44bc9ebc0e9f8fd32a68347922ac39cf.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "69b9057f521320c315dc9b98206300c0b7c85bea77860657c1bfcba43491513b", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9788dd97217964965b6e50aa82ff1ca109987ae5347b696142cd39771ea4be54", "generation_seconds_batch": 16.324127857922576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.7205564975738525, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/69b9057f521320c315dc9b98206300c0b7c85bea77860657c1bfcba43491513b.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "6a269cdb1a06c10df2190a1dcd79e76504d21a6640af1931c4383e9fcc527e5a", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3820eab55d724a32d7e108ddaa11c6273c88c2a1fb53c7daa35d26cc0cd4d93a", "generation_seconds_batch": 17.311734304996207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.481858015060425, "target_z": 0.9687500329360282, "ce": 6.284721851348877, "genuineness": 3.9801597595214844, "identity": 0.6744006872177124, "hypothesis": "ik heb het kleine doosje gevonden het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6a269cdb1a06c10df2190a1dcd79e76504d21a6640af1931c4383e9fcc527e5a.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very warm vocal delivery. 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": "230a2eb1c441edc034c84b229c61db31e12ea65056e19623c3a3594463cb8268", "generation_seconds_batch": 18.64826293999795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.4515180587768555, "genuineness": 4.017816066741943, "identity": 0.12934087216854095, "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_WARM_reg_high/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "6af3de80a44755eba53b4a14e0e6a20d9a7cade33a47aab43ac2aeb0611189ae", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "05728133288cee8d1d5030e304d05abde94f00d68dd00c82057caaae4a60e4cf", "generation_seconds_batch": 22.740851672017016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.260893821716309, "genuineness": 4.137060642242432, "identity": 0.21270322799682617, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6af3de80a44755eba53b4a14e0e6a20d9a7cade33a47aab43ac2aeb0611189ae.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "6b13c7c032af37647b1dcaffaee265489264f5f550e364f3f7bd1a346925482d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6837e17ad21e97dd1edce98b122087700c177f223f54e55b1db829acf036ace5", "generation_seconds_batch": 28.0569482780993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.165581703186035, "genuineness": 4.375550270080566, "identity": 0.6181637644767761, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6b13c7c032af37647b1dcaffaee265489264f5f550e364f3f7bd1a346925482d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "6b33e8760060934d11d51d270972d073056135e67bf946555d6a3fcea13ba387", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1e78cb98384bd2e5af659262fd66108138e6f6ef083fc634b5787c786a7b6898", "generation_seconds_batch": 25.501212641014718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.078212261199951, "genuineness": 4.494794845581055, "identity": 0.6800339221954346, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6b33e8760060934d11d51d270972d073056135e67bf946555d6a3fcea13ba387.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "6b3cadb536aab235a367eeea1c4a7a633fdafea57b51d4efa8cc781dc3b9021c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a68de3bca8a19dc9b9c81378ff4f8d05bb281fd7216c36853217d091ec29f7cc", "generation_seconds_batch": 28.754313410026953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 5.129063129425049, "genuineness": 4.6203155517578125, "identity": 0.7494580745697021, "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_WARM_reg_high/6b3cadb536aab235a367eeea1c4a7a633fdafea57b51d4efa8cc781dc3b9021c.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50cf4afe7a0b3f58f6236cd7c68b0376ff1bd273124c4fd2d7d66795deef703f", "generation_seconds_batch": 15.486682903952897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.308549880981445, "genuineness": 4.6203155517578125, "identity": 0.6275525093078613, "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_WARM_reg_high/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "6bde6338f0279eda17cf04b5e5ddace21c1c85c46a24c2e81caaa0d02f918b7d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4e979fbc95261aaf2ab7680ab86cd1e2fda44f909a676ffa21006d7ee22ce335", "generation_seconds_batch": 24.363136643078178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8565244674682617, "target_z": 1.3828125576380494, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.37957364320755005, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6bde6338f0279eda17cf04b5e5ddace21c1c85c46a24c2e81caaa0d02f918b7d.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "6bf9ac80b059fe8420fc93f5ca0b8330fc53a84e33bf3aada171030f99df977d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1d153bb29ecdca25299458ec95a4d768849e47feb78231bd2d60e11b38c384d6", "generation_seconds_batch": 17.883928347961046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8141093254089355, "target_z": 1.335937475297979, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.7600207328796387, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6bf9ac80b059fe8420fc93f5ca0b8330fc53a84e33bf3aada171030f99df977d.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "6c0d55c8bee07a70577c4fe68d69c9ec7dea94c27475b8b167ced1be51c5522b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d07db8f4d9088e096a93914aa80c7de35c974ba7bbd6aa1e6571f1672539cc98", "generation_seconds_batch": 23.931804548949003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7716941833496094, "target_z": 1.2890623929579084, "ce": 6.443574905395508, "genuineness": 4.105680465698242, "identity": 0.6241143345832825, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6c0d55c8bee07a70577c4fe68d69c9ec7dea94c27475b8b167ced1be51c5522b.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "6c86df3631c421b5e3c75c79cfb392417a095b0a3528566d662e0be8b0069e02", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2e7e8cd12790ff4d5b78cb6e9750388c2b11ebfb7cda64c9df48cbc9ddc2e9c7", "generation_seconds_batch": 24.336334924097173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 5.990842819213867, "genuineness": 3.810706615447998, "identity": 0.5515741109848022, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6c86df3631c421b5e3c75c79cfb392417a095b0a3528566d662e0be8b0069e02.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "43ed121395e51b0fb7e4a48437c26d5419de825c4393ebe31cb559a77bc3b108", "generation_seconds_batch": 17.274609473068267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.7320642471313477, "hypothesis": "De weg gaat linksaf naar de brug, loop verder, tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6c8d1cb1159440aca706b72e34f7743340d2b66a066c9c97214892bdbd0a3768.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "6cac894eb5f07fbecd8d2f7c31ccc5a7fa58f813c1665258d564654b5597db6a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9a5547cc70104705ebfa45aaa4cd95f07ed8938463da892f8892f4bc19a6d318", "generation_seconds_batch": 22.097469568019733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.096876621246338, "target_z": 1.6484376729141481, "ce": 6.054384231567383, "genuineness": 3.9425034523010254, "identity": 0.5911078453063965, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6cac894eb5f07fbecd8d2f7c31ccc5a7fa58f813c1665258d564654b5597db6a.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "6caff9cbeb0576d62abb86fcc21b883ae627bbba8cd75a403906f7c424c548c6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b01cecb02efb6ba52a29e904c79b9dba035aa1edab8eea3269adedfaef01cb30", "generation_seconds_batch": 17.561375727993436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471254348754883, "target_z": 0.9570313940951235, "ce": 6.38797664642334, "genuineness": 4.193545341491699, "identity": 0.5513814687728882, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6caff9cbeb0576d62abb86fcc21b883ae627bbba8cd75a403906f7c424c548c6.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2b8d8a6fe254e6cbbebc93759473a74a08abf776fcc5014373d04c3b70f37843", "generation_seconds_batch": 25.65294576191809, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.41180419921875, "genuineness": 4.18099308013916, "identity": 0.45882540941238403, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6cc3a8a39daca47afb0c5017ef6b5422fbbdc0f366ad3b4ccbe9ca13167218d9.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "6d1626b7e436e1a700f7f0c2399e650d49a809f585471c3104ef8c4bbfc81d35", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "874ba1fcd13b914146e080121c9c2a09ddcff9bbe4f025d750f9cd6d73dd78fd", "generation_seconds_batch": 23.959351639961824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.133810997009277, "genuineness": 4.030367851257324, "identity": 0.3845621943473816, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6d1626b7e436e1a700f7f0c2399e650d49a809f585471c3104ef8c4bbfc81d35.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "6d91ea3d245c718b4f2abe60947e211c7974a64656682547a5c2263cf23d9ddf", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f63620cfddf592c17a63f0b012c3e719229164e91dab548ef671b49468f44769", "generation_seconds_batch": 19.67038721498102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7151408195495605, "target_z": 1.2265624588299646, "ce": 6.260893821716309, "genuineness": 4.457139015197754, "identity": 0.602841317653656, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6d91ea3d245c718b4f2abe60947e211c7974a64656682547a5c2263cf23d9ddf.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "6d9c7f09906553c13cf45fcae8c399243cc4550c1471ad9a4bf527fae424866d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f64508e14ac397919fe2713059362e371d9e1f25b320abb11ec3d1ad8d40091a", "generation_seconds_batch": 15.949976561008953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4959964752197266, "target_z": 0.984375148212127, "ce": 6.324435234069824, "genuineness": 3.553389072418213, "identity": 0.31026360392570496, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6d9c7f09906553c13cf45fcae8c399243cc4550c1471ad9a4bf527fae424866d.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "6edfd894018f79b9dc49cc685c2789f370e41773f4d7146978fdb21d51e792e5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "eb0b013775768d53fa347760b1aabd8849449320153ed2802a84cc97078c9e14", "generation_seconds_batch": 24.067220276920125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2311906814575195, "target_z": 1.7968748188518449, "ce": 5.77639102935791, "genuineness": 4.005263805389404, "identity": 0.7007530927658081, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6edfd894018f79b9dc49cc685c2789f370e41773f4d7146978fdb21d51e792e5.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "6fd5accbb87d491c48d46a2379901087f5e02a741b4d5f7c1ae4b95af8403743", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4a6240753e649d324ec12e0fdcf3bb858f2deeb8cf4d54c6d3c88411c632311e", "generation_seconds_batch": 18.06516317697242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.09409761428833, "genuineness": 4.65797233581543, "identity": 0.7935984134674072, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/6fd5accbb87d491c48d46a2379901087f5e02a741b4d5f7c1ae4b95af8403743.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "704893e5e79c8c4f6896628bd4e2e67c82f7c2fdbc3cdf4207498baef1f52ea6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ab0d3459b50569c6c39f992d3b310e7e1102f940a2820951d6f91ec89190ef51", "generation_seconds_batch": 23.19145959895104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.4912309646606445, "genuineness": 4.306513786315918, "identity": 0.637887716293335, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/704893e5e79c8c4f6896628bd4e2e67c82f7c2fdbc3cdf4207498baef1f52ea6.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "7065454bee073544e58e2cf504673bc009665d7fd333e9469735589b96d341ca", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fb6f9b14f61eab6225a09245c333ace95ce6462fa9c84d153617edbb806a0c75", "generation_seconds_batch": 24.12149797496386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.229123115539551, "genuineness": 4.017816066741943, "identity": 0.2765369415283203, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7065454bee073544e58e2cf504673bc009665d7fd333e9469735589b96d341ca.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "70820c0a7db29244c710d4e00925146c1cad340f7b5a09126e2a6c54e93cf2d8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4d7e85894cce174bccb149156f251de7f63df2c25c4db1dba1fab48cb8707c1e", "generation_seconds_batch": 17.55741760507226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4182353019714355, "target_z": 0.8984374094259224, "ce": 6.197352409362793, "genuineness": 4.413206577301025, "identity": 0.5342847108840942, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/70820c0a7db29244c710d4e00925146c1cad340f7b5a09126e2a6c54e93cf2d8.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "70b8a234f1b471773e194bdb4c7b559f6f99f96e4d019a63000111e74bc4be5e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8585c1e9a276b6ff50bf7a0f73927c0c1d6801f8ae4abe1657f7535e6d8a642f", "generation_seconds_batch": 22.31618132593576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 5.974957466125488, "genuineness": 4.7458367347717285, "identity": 0.8416627645492554, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/70b8a234f1b471773e194bdb4c7b559f6f99f96e4d019a63000111e74bc4be5e.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "70cc4245254fd91248fe27c8062972821ec1fdb52980de6793827e315637ac9e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "924acb6c6c79bdabe0402ca460fdb61fbe09486261f85e49948e2b525dbbdc7f", "generation_seconds_batch": 23.7681716469815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.173524379730225, "genuineness": 4.262581825256348, "identity": 0.5834348797798157, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in de kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/70cc4245254fd91248fe27c8062972821ec1fdb52980de6793827e315637ac9e.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "70f5bc798639ed7d98c4f3f72004ad18ef280c027403f8da4949a04a874e91d3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6620377a68c25a0e39a3931ba803a368760dca98c8bf1d7b12e66d64cdfabbb9", "generation_seconds_batch": 17.66451627900824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.252398490905762, "target_z": 1.820312623510106, "ce": 6.1020402908325195, "genuineness": 4.231201171875, "identity": 0.5316436290740967, "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_WARM_reg_high/70f5bc798639ed7d98c4f3f72004ad18ef280c027403f8da4949a04a874e91d3.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "714e84217a95fa0fdfd03ab84f43546e7eee049024d64f3bbe037648441fc4e5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "bc7ea48b12f27248f329c4b3a6c6dc50980e3f6a0246d5c8d1ade51d8ea028a7", "generation_seconds_batch": 16.5599813190056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.213237762451172, "genuineness": 3.9048471450805664, "identity": 0.29377129673957825, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/714e84217a95fa0fdfd03ab84f43546e7eee049024d64f3bbe037648441fc4e5.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "7248b4f7d944ed90194ea979fc168100ad6fa79b2fe4ba8cd23bf3f4d164c326", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "8836a55fb57efda33510e570f4e4335f81d88e68ad9599b78a91f71493477acd", "generation_seconds_batch": 24.779135588090867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.252951145172119, "genuineness": 3.9676074981689453, "identity": 0.47012293338775635, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "wer": 0.0625, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7248b4f7d944ed90194ea979fc168100ad6fa79b2fe4ba8cd23bf3f4d164c326.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "7267ec3a3b75647efd5bfa5728c8a05f72908e10662d9b4d8efc755d464a0217", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "dd428ded1d9e7928ab0322e585df8167dcea17d07ec791d8b388e89e5c912ee3", "generation_seconds_batch": 17.802628564997576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.133810997009277, "genuineness": 4.469690799713135, "identity": 0.43547189235687256, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7267ec3a3b75647efd5bfa5728c8a05f72908e10662d9b4d8efc755d464a0217.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "7280a36f3bf3afac54eda7bb4fbb4ee96231dd3baa6eedec801c6160d6cf3d93", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "4d0fc50740fd0c01c1a9014931c1198eeb78fcced33baa8b49824871d539c4e3", "generation_seconds_batch": 17.076868982985616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.2767791748046875, "genuineness": 4.375550270080566, "identity": 0.7576278448104858, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7280a36f3bf3afac54eda7bb4fbb4ee96231dd3baa6eedec801c6160d6cf3d93.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "72de55827f42df43f83ec08b22d35384358f266aefbd2aad7a8e1687fb84e1de", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c6cb103e7e42461f72a2d7dc84abb34e767401027f8d9b713c1d9f5cb749573b", "generation_seconds_batch": 24.12173402903136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.24500846862793, "genuineness": 4.068024158477783, "identity": 0.4266279637813568, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/72de55827f42df43f83ec08b22d35384358f266aefbd2aad7a8e1687fb84e1de.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "72f94b36842e723d9e7fe75a4bafec13c47730b1006839593733429d21f8f238", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e184b68f7aed607f68ed99fd4d47e5888cf805c6df50bec812764b07d183a46c", "generation_seconds_batch": 16.04885371995624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.348263263702393, "genuineness": 4.457139015197754, "identity": 0.39439862966537476, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/72f94b36842e723d9e7fe75a4bafec13c47730b1006839593733429d21f8f238.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "73bf4d8eebe6b207d675c903720bea6d2a82eda7c7dc2b1d4d71c9692d7fa7c5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6cd4966279e8f23abfa4c045f19c4d8adb12c61ea41c208966b03097b474b69b", "generation_seconds_batch": 22.642339496989734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.768048107624054, "hypothesis": "J'ai trouvé la 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_WARM_reg_high/73bf4d8eebe6b207d675c903720bea6d2a82eda7c7dc2b1d4d71c9692d7fa7c5.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "73cade006d3dcfd97a552de63d3a44dc66fd5f238fc226dc6402bc8fd0f76286", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "32d2c63077ecb1b1624478b54fe50c2314d80935d62397d86a6e8040ce908d99", "generation_seconds_batch": 24.791064786957577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.348263263702393, "genuineness": 4.105680465698242, "identity": 0.32199615240097046, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/73cade006d3dcfd97a552de63d3a44dc66fd5f238fc226dc6402bc8fd0f76286.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "743d1df3a8ec3284615f61f624d54b5bd4a009e53e09b3e4f9733eeee7a366e1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9d3cdc86b782e602e017aae134c4cd094bbd1d2dd30a3ed1ec7f30e4de5842eb", "generation_seconds_batch": 24.35213533404749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.292664527893066, "genuineness": 4.080576419830322, "identity": 0.3062911033630371, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/743d1df3a8ec3284615f61f624d54b5bd4a009e53e09b3e4f9733eeee7a366e1.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "746d0e4e57d8735b1b58f4887fa21ebb41a5ab9551189458187b7600bd426631", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "09d80bf00ce03e092da1f8aaf00c036aeaac4086c3ce7b6f339d71db731fcc90", "generation_seconds_batch": 17.307742376928218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.181706428527832, "target_z": 1.7421873106178378, "ce": 6.300607204437256, "genuineness": 4.287685871124268, "identity": 0.46454504132270813, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/746d0e4e57d8735b1b58f4887fa21ebb41a5ab9551189458187b7600bd426631.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "74aea914dde204d2041770a2895e0bad7300fe994e0541075dfd1a6f9eea1a24", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6906c793045f6030a2144d2a1d3ef9cb86e2057d9afaf1ce51cb8dc023012d71", "generation_seconds_batch": 22.67714912991505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.340320587158203, "genuineness": 4.030367851257324, "identity": 0.14220212399959564, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/74aea914dde204d2041770a2895e0bad7300fe994e0541075dfd1a6f9eea1a24.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "74bb03650914825374b858b78f0f25eacdb872fe437cb0d8b5c8eaed5480a6fd", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a79203bb8c3367bea6cde1c6995b4a764aa1af096f34b69ec4a7a862944d4e80", "generation_seconds_batch": 16.143981837900355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.205295085906982, "genuineness": 4.074300289154053, "identity": 0.540619969367981, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/74bb03650914825374b858b78f0f25eacdb872fe437cb0d8b5c8eaed5480a6fd.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "74edecf96f60d4e5320db9804d4f2618ba2bcb17265537008b104f40428efe0d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "dd0ec08406bbdfc845fa4cd2104eca512dcf26e80ee3e258102fdb5ce6a3fbab", "generation_seconds_batch": 21.40773731400259, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.2767791748046875, "genuineness": 4.017816066741943, "identity": 0.3777717351913452, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/74edecf96f60d4e5320db9804d4f2618ba2bcb17265537008b104f40428efe0d.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "74fa5ce85b0d1903a2ee4930e902fb8e0139436cdf5fa52be9753ac97e7eb046", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fd98ef9bc2f7ecbd6d8af18b1012eb5174768060c37619a11af32be5e01ecb27", "generation_seconds_batch": 20.859213036019355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.404097080230713, "target_z": 0.8828125576380493, "ce": 6.4912309646606445, "genuineness": 4.432034492492676, "identity": 0.6320558786392212, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/74fa5ce85b0d1903a2ee4930e902fb8e0139436cdf5fa52be9753ac97e7eb046.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "75e7a95acf8034e1dc587f4e332d679a3be5d8aad7cee80d5719a6d30cdc6284", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f9399df9e372d5f74711604bea37ad37663ca83570dc7cbf436b940635f5991d", "generation_seconds_batch": 22.061112171038985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.054461479187012, "target_z": 1.6015625905740776, "ce": 6.372091293334961, "genuineness": 4.362998008728027, "identity": 0.3271886110305786, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/75e7a95acf8034e1dc587f4e332d679a3be5d8aad7cee80d5719a6d30cdc6284.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "760c26fb065be68f4f7f0aa163ae3cd28fcdfe223f2c8c2694a6fb461121c44b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "025072f75c221b02f032cc74b5a7ea1a472f5d38ddba0e342c088146946a5acc", "generation_seconds_batch": 23.29593155893963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8141093254089355, "target_z": 1.335937475297979, "ce": 5.982900142669678, "genuineness": 4.256305694580078, "identity": 0.6485561728477478, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/760c26fb065be68f4f7f0aa163ae3cd28fcdfe223f2c8c2694a6fb461121c44b.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "764acbd5103c456f396a3ddf739cbed8e2ccb45dceacbd2e534495771f328c8d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a019a68257f4780cb269a24ddccc6f7f5e6260c4068f1f0b92ecad9c394f3595", "generation_seconds_batch": 17.67146022093948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.34158390760421753, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/764acbd5103c456f396a3ddf739cbed8e2ccb45dceacbd2e534495771f328c8d.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "768236a6bd5becabf64452bf785d75a071e3949b80e6a61b9c1621177c0c0ce2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5620912c4ed7b9f653f405d3c3ff2987d50faf1c337dc7f0408f1b7dbabc3c26", "generation_seconds_batch": 23.666640614974312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941354513168335, "target_z": 1.4765624588299646, "ce": 6.173524379730225, "genuineness": 4.3881025314331055, "identity": 0.6071649193763733, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/768236a6bd5becabf64452bf785d75a071e3949b80e6a61b9c1621177c0c0ce2.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "76aaa81870976414a94e3ae41720ddb17012dd8ef1a01af7a9c7e5597f3d863e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9d9ce688714a2503ccdd49ec6de5261c84daa88e0f611cbf9a5cc00a47da5f43", "generation_seconds_batch": 23.27456595201511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.386713027954102, "target_z": 1.968750296424254, "ce": 5.609594821929932, "genuineness": 3.9425034523010254, "identity": 0.6415785551071167, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo es.", "wer": 0.13333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/76aaa81870976414a94e3ae41720ddb17012dd8ef1a01af7a9c7e5597f3d863e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f1f2dcbe004998165c5d4279d6858fd34752f31532147972cdb6aea24c151223", "generation_seconds_batch": 18.088549594045617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.427689552307129, "genuineness": 4.193545341491699, "identity": 0.716069221496582, "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_WARM_reg_high/771eb0f298d0e962e3e9dc7ee00c3926f715d41d3f4c43f41d6fa45d6f1e2b15.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ce46bd034d83bf17685f27d6182b61a59030418698079eb6360bf18b347c4321", "generation_seconds_batch": 24.29494415107183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.006728172302246, "genuineness": 3.9550557136535645, "identity": 0.6627336740493774, "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_WARM_reg_high/774a40a75fa4637a280e8f908935c10bf2981436ef405866661dd80391faea7b.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "77932428c3b18d6191f48171c6714e69092b690bd6979a1670f1f58c4a1645ce", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9ff67b8d410d46f613045890f6cf51194fa4f051dc48fd14c1137e8ae0fb612d", "generation_seconds_batch": 21.03391462098807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3616819381713867, "target_z": 0.8359374752979788, "ce": 6.300607204437256, "genuineness": 4.18726921081543, "identity": 0.4500150680541992, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/77932428c3b18d6191f48171c6714e69092b690bd6979a1670f1f58c4a1645ce.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(very warm vocal delivery. 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": "544d2c00baf0a653a0dbfbfe2d1abf1ca8df955e547296e891105e7fd76ef6b4", "generation_seconds_batch": 15.925009939004667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.7642630338668823, "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_WARM_reg_high/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "77a64c3357feed73f23eefcfbfd53b7848c3dbdd1c32ea9521cd554b9d06f495", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "68e39c3a75f58c022812bbcd9b596761e9c92074f09d4187af6391733486c515", "generation_seconds_batch": 21.662563862046227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.5723625421524048, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/77a64c3357feed73f23eefcfbfd53b7848c3dbdd1c32ea9521cd554b9d06f495.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "77d91b6dac73fb5122bf270f981d3336de5dc557856de664b1cdd9d0c650fb18", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0a87c522dd6957524c164f078da328b41d76cf271043e9bcb00b6b8d6aea1b9f", "generation_seconds_batch": 22.080510414903983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.146360874176025, "target_z": 1.7031251811481551, "ce": 6.403861999511719, "genuineness": 4.369274139404297, "identity": 0.7111920118331909, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/77d91b6dac73fb5122bf270f981d3336de5dc557856de664b1cdd9d0c650fb18.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "782b46732a209ce0acdca51b26b96c8127d45d21cbac9c4f09c0d938143a39f5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2c83900b16664f18e6a93f9324416fe51cc660f6cf4ea9572ac22ba7df13f12d", "generation_seconds_batch": 24.09494421002455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.260893821716309, "genuineness": 4.068024158477783, "identity": 0.2635457515716553, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/782b46732a209ce0acdca51b26b96c8127d45d21cbac9c4f09c0d938143a39f5.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "786e2fb7430fb50b3d262c2992cc27a9a28f2a2377c0efb0234c5576c26f6d08", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "fe1ebdb100dbf88cc85f2cdcfb2719577a18235364c067c792c94295521e130a", "generation_seconds_batch": 18.909976088092662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.372091293334961, "genuineness": 4.080576419830322, "identity": 0.4183011054992676, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/786e2fb7430fb50b3d262c2992cc27a9a28f2a2377c0efb0234c5576c26f6d08.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "31761dc91cad1e9a79dd2a2631e963269367fa293f78d428ac7b0ec4be795bea", "generation_seconds_batch": 18.061762645025738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.655341625213623, "target_z": 2.265625115276099, "ce": 6.205295085906982, "genuineness": 4.607763767242432, "identity": 0.35685864090919495, "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_WARM_reg_high/789926d4a04c82cdaf5e682f7270db538413f91693b7f24a2bac1a804e6d4c7e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "296bb7ecf3ed5f426c73108e91dd13e65776db788b6b8ea97ecc8f0719e06599", "generation_seconds_batch": 21.21823633206077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.983769655227661, "target_z": 1.5234375411700354, "ce": 6.2767791748046875, "genuineness": 4.105680465698242, "identity": 0.7780259847640991, "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_WARM_reg_high/7920b7769c59ea0733607fe890db951a548946751217941a4d289d8050e5c2d9.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "7933a488a16c051e316c91f7e79942ced8a8cc56dce31200bedf6d7e86f7ccc0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "38dcc24ce480fd397b503c4ddfef2ddb1ac0620cb8f7687e3c524888dcd343d7", "generation_seconds_batch": 18.51516955893021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.41180419921875, "genuineness": 4.331617832183838, "identity": 0.4418462812900543, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7933a488a16c051e316c91f7e79942ced8a8cc56dce31200bedf6d7e86f7ccc0.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(very warm vocal delivery. 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": "bab683db244e4059b5a6670e583dce8b591a6187bceaa3ad1f0b8431e09efa43", "generation_seconds_batch": 17.434214334003627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7716941833496094, "target_z": 1.2890623929579084, "ce": 6.1894097328186035, "genuineness": 4.469690799713135, "identity": 0.4909585416316986, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c7ab5aad43608a25999440365d59c0dc934c50aad261634cf99bca2073a4160e", "generation_seconds_batch": 23.325941888033412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.1020402908325195, "genuineness": 4.6203155517578125, "identity": 0.5857247710227966, "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_WARM_reg_high/7961b7b871db3fd7da0d44df26239925e852e22d878cd142eaee4cf8e6a216a9.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "797d213362f55e083ff48f1c33fdd8e07f62facbc30f38571a96739c9c694f5e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b76846b5afcc47c26076378c976ba3421a8621a48249845d3172fbb4395164c3", "generation_seconds_batch": 17.588252476998605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.040323257446289, "target_z": 1.5859377387862046, "ce": 5.262103080749512, "genuineness": 4.369274139404297, "identity": 0.7826414108276367, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/797d213362f55e083ff48f1c33fdd8e07f62facbc30f38571a96739c9c694f5e.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "79f5d2adef08735ba19d5f32ec34fa265a3cf041ad6284614e846f2c057fe674", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b3b68e1d9665e1b0e850c589a520f75f9ca19de3336ad144bae73cda9f5ad115", "generation_seconds_batch": 20.275813754997216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.332377910614014, "genuineness": 4.362998008728027, "identity": 0.5585896372795105, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/79f5d2adef08735ba19d5f32ec34fa265a3cf041ad6284614e846f2c057fe674.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "79fa61e19c7f5191e536cd4b66985b74960b85012eec81198cb129fde13040d5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c052c4912bdcbb4a25ad7904b1c223b1d69688eb11c30852d9fbe7a71b219b62", "generation_seconds_batch": 16.70772501709871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4182353019714355, "target_z": 0.8984374094259224, "ce": 6.078212261199951, "genuineness": 4.18099308013916, "identity": 0.6598290801048279, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/79fa61e19c7f5191e536cd4b66985b74960b85012eec81198cb129fde13040d5.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "7a12e375e6b5071b95c2c0b1385b1479187a90241badfd3c0de4ec7bda9331e3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "623238d094242230d1b0ca7221fbdbf13bd76dd092b69151a98031eff98b93d5", "generation_seconds_batch": 24.051191041013226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.125153064727783, "target_z": 1.679687376489894, "ce": 6.0305562019348145, "genuineness": 4.413206577301025, "identity": 0.8154392242431641, "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_WARM_reg_high/7a12e375e6b5071b95c2c0b1385b1479187a90241badfd3c0de4ec7bda9331e3.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "88e23d6c1f7c717e269a4cef1af11e7f1baf4dc6d2d6250798855a62af388dc4", "generation_seconds_batch": 23.86574003007263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898939609527588, "target_z": 1.42968763997812, "ce": 6.419747352600098, "genuineness": 4.287685871124268, "identity": 0.2016235738992691, "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_WARM_reg_high/7b29a2517ec2d5fc796f11712be3c2bb36075ccd1ea94655f39871d2c9463cf9.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "7b7bd15f144bc281950ca2b583f18f4b3c13911cd6d76bcd78d629fb934d33d6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a82c469ffd2f6fc85c426e6af0a087b095eb40a3715a985a9e3db0a5bdaaad23", "generation_seconds_batch": 22.750352279050276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 5.831989288330078, "genuineness": 4.331617832183838, "identity": 0.7461780309677124, "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_WARM_reg_high/7b7bd15f144bc281950ca2b583f18f4b3c13911cd6d76bcd78d629fb934d33d6.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "7bc7c19ba533c34045f3f4f4da78695747ab4c8ba46079a8077364964bbe2ecd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "275c583a3cb517051be4495d52a9b57913aa02c918d4f1d5ada7962e6fb22852", "generation_seconds_batch": 22.67714912991505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.405903160572052, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7bc7c19ba533c34045f3f4f4da78695747ab4c8ba46079a8077364964bbe2ecd.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "7c44709c81cda262db74beef865a42005147bcdd1af95e2b5df2aca345c6a118", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7d702b55c200839918629ba4737e6667996fd8402208f881bdb0b86f46ca73c6", "generation_seconds_batch": 16.91247162304353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.197352409362793, "genuineness": 4.074300289154053, "identity": 0.7076414227485657, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7c44709c81cda262db74beef865a42005147bcdd1af95e2b5df2aca345c6a118.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "7d078da53aaf1ff826bdaf7e3153c7feb375fb1ad8caa4d8803570132fb6da95", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "29afe268db62bcc5f48469013124024f779d59caadd7a2dacf326f20f308b300", "generation_seconds_batch": 17.94504861102905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.372091293334961, "genuineness": 4.406930446624756, "identity": 0.5875054597854614, "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_WARM_reg_high/7d078da53aaf1ff826bdaf7e3153c7feb375fb1ad8caa4d8803570132fb6da95.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "7d7527bf489ac2088212e9035e573432158a521c5a516ab104917704b9748ede", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2f6c43e5864ddec70f1f6c79c9666b555a442e3031ad4651b9c5ab61ea3bcef2", "generation_seconds_batch": 16.483222765033133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.221180438995361, "genuineness": 3.9801597595214844, "identity": 0.7264739274978638, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7d7527bf489ac2088212e9035e573432158a521c5a516ab104917704b9748ede.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "7db0fe2eb6d10cce8ac1aae06140509af543132adf48f8ea1056ae7e2387ddd2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3b62839b0da9523d0d7399576dbe6cbb69095a503010c41ad766ff3cc857a290", "generation_seconds_batch": 17.039087084005587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.082737922668457, "target_z": 1.6328122941498235, "ce": 6.173524379730225, "genuineness": 4.65797233581543, "identity": 0.6490871906280518, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7db0fe2eb6d10cce8ac1aae06140509af543132adf48f8ea1056ae7e2387ddd2.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ff54d3593ddbee3bd43a9bc4a8e3f17a9230cc45d3d55194fb65ca7660447ae5", "generation_seconds_batch": 15.340972983976826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.828247547149658, "target_z": 1.3515623270858519, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.7597327828407288, "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_WARM_reg_high/7dbf8ddd00a7d1080fd6c5f63c726e476f85afb120fb2b37a4274d78ea50e162.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "7dcb884c4c74185f32be48133beb9ac9bad4e164d6e88cf91dd4a85095f22a69", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ed28d9ebfbeaeaf67388e499368eeb5bbbb6729c00024b20141fd1c99fbe79d6", "generation_seconds_batch": 22.57222345902119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3192667961120605, "target_z": 0.7890623929579083, "ce": 6.078212261199951, "genuineness": 4.306513786315918, "identity": 0.679025411605835, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7dcb884c4c74185f32be48133beb9ac9bad4e164d6e88cf91dd4a85095f22a69.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "7e325b0f8edbce89f82b2b81bb4e3fed7762dd08320ba13655be5733a663314e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "3346cf83d6aa0b7a029e90f20533c3f4eec789b4ebf77268f3604e6c676eba4b", "generation_seconds_batch": 25.15820707299281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.173524379730225, "genuineness": 4.086852550506592, "identity": 0.23844827711582184, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7e325b0f8edbce89f82b2b81bb4e3fed7762dd08320ba13655be5733a663314e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "259033904b16460d25052b33ba7fc1b1c2573a8ec6813bb3f580f20e11d0ea8e", "generation_seconds_batch": 24.02243950893171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344297409057617, "target_z": 1.921874687107732, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.7168647050857544, "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": 0.120000000000001, "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_WARM_reg_high/7e3dfcb8b511aefa26ed96b60b7f8a84e44c84eac24e6bb8a4b4a370d8af48e4.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "7e5080217e97da9ce80fb29a03153786953acb432f66c2498b9ea068d72e101e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1b0962959eaea47ddd2d1215ca497f69699e3cf22fc9e76a9f9f4743523a17aa", "generation_seconds_batch": 28.195202487986535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6303107738494873, "target_z": 1.1328125576380494, "ce": 6.23706579208374, "genuineness": 4.444586753845215, "identity": 0.6737489104270935, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7e5080217e97da9ce80fb29a03153786953acb432f66c2498b9ea068d72e101e.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "7e6637e8e848b027ab8806f472d34ebe1b752c3ec71888ffffb23d078be99020", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e6c4e7618fbef683981cc4884ece10885779b5a628f1a118fb186650c1cad1cf", "generation_seconds_batch": 17.561375727993436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.372091293334961, "genuineness": 4.256305694580078, "identity": 0.6756417751312256, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7e6637e8e848b027ab8806f472d34ebe1b752c3ec71888ffffb23d078be99020.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "7f27c1d361bf8c5accb7c869ea339b5956dc780ec3b7b89163d761c9240119f5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "23fcd9bbf534a03886fa09ce9afdda3351fb88d781dbb65af1725777a03817d4", "generation_seconds_batch": 23.508138348930515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.290990114212036, "target_z": 0.7578124258939365, "ce": 6.213237762451172, "genuineness": 4.1998209953308105, "identity": 0.5569080114364624, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7f27c1d361bf8c5accb7c869ea339b5956dc780ec3b7b89163d761c9240119f5.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(very warm vocal delivery. 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": "0b4223170e9592675f8b646bf4decf5769c02b4a92ccd092c2ae6b6bf11e8b8b", "generation_seconds_batch": 17.584979100036435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.348263263702393, "genuineness": 4.030367851257324, "identity": 0.3289608955383301, "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_WARM_reg_high/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "7fcc6b4afc5f6c550db3930e2187d3c2a34f85f27365361e885848a9fc1d6642", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b44c0a48b4be09bac2ceebdc83453472698f72e52de8b66a794c06f6c4b72ce3", "generation_seconds_batch": 22.280806713039055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3301591873168945, "target_z": 1.906249835319859, "ce": 6.125868320465088, "genuineness": 3.998987913131714, "identity": 0.816164493560791, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/7fcc6b4afc5f6c550db3930e2187d3c2a34f85f27365361e885848a9fc1d6642.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "61f1c762614bc8f3d4cc228c916732c9c72109462a5ddda2e55f89ffff3e4d4d", "generation_seconds_batch": 16.645318311988376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.586543083190918, "genuineness": 4.068024158477783, "identity": 0.3903365135192871, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/80761b70e06b85a45b410aa8cde5b04b9cdadc7f41fc612fc05a47b4c4360f52.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "80ad0e483d44ec3e80e8dff9f5e1ec8deb3f6f159b79e6ffe008057018d095b5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a5e41412392218a8ae34438a039521e3827049903e2083721158e25f554e9785", "generation_seconds_batch": 15.663766894955188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.197352409362793, "genuineness": 4.281409740447998, "identity": 0.7683509588241577, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/80ad0e483d44ec3e80e8dff9f5e1ec8deb3f6f159b79e6ffe008057018d095b5.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "80afbd525512b86abbfd1cdf567727aec19aa42f5aa1909216fdd844aa1bf952", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ed8b9e8c6837360a987742d42cb322cfd74bf1e5d667735f2ed9ed525b4f58f7", "generation_seconds_batch": 23.718183303950354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.459460258483887, "genuineness": 4.331617832183838, "identity": 0.5408486127853394, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/80afbd525512b86abbfd1cdf567727aec19aa42f5aa1909216fdd844aa1bf952.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very warm vocal delivery. 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": "0f197b0b500bbf09cdbd6a5c3885acce1f678b06c0a63aa1e4846420d6f3f818", "generation_seconds_batch": 17.594193930039182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.078212261199951, "genuineness": 4.2751336097717285, "identity": 0.6081362962722778, "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_WARM_reg_high/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "80d149c60a6d968e0e5643ebfcb46b4db5a5221f006b2514a07be8f7ab611536", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d0da86ca51fff9481c9cf0c23de2fc9a5448b9a407ae439e43cc2cd2632c139e", "generation_seconds_batch": 17.483919998048805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.443574905395508, "genuineness": 4.231201171875, "identity": 0.6455590724945068, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/80d149c60a6d968e0e5643ebfcb46b4db5a5221f006b2514a07be8f7ab611536.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "80d96aaad0b46b8f30e5d443944daa0f4d6a28dcb7829449f17ea90c43b40de2", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b97cb60301fc912c3a4088b331827ba3cefc7bd365a136226b2db3db106d5800", "generation_seconds_batch": 17.96937737299595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.170814037322998, "target_z": 0.6249998682558872, "ce": 6.0305562019348145, "genuineness": 3.7793264389038086, "identity": 0.260539710521698, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/80d96aaad0b46b8f30e5d443944daa0f4d6a28dcb7829449f17ea90c43b40de2.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "810b90e424a432df742dd1fa61f8119378fc13c38bd73b6c6d675c8281741ad9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cc86654097f44950ff3b6021d7b80fc73694a8c3575e34d0b0c3eb86d1318d40", "generation_seconds_batch": 16.684433295042254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.40220382809638977, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/810b90e424a432df742dd1fa61f8119378fc13c38bd73b6c6d675c8281741ad9.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "09fa585d256d63663871dd7a5fba435697811e8d133e34c486f653903010aa3d", "generation_seconds_batch": 23.86574003007263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.419747352600098, "genuineness": 4.18726921081543, "identity": 0.34323060512542725, "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_WARM_reg_high/81381944ee0c81732e50b513fbaa2087d0491bb304e65dd2aa7065e56f163314.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "8224d5d7607cb98531ce56b4db17bee4240c6fb76652ee0a27b7e217474f2cf2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "92411cbfc98597a1db1b0be3c0bf50741db736399c194a7f738ab392532ec498", "generation_seconds_batch": 28.195202487986535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.6345603466033936, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8224d5d7607cb98531ce56b4db17bee4240c6fb76652ee0a27b7e217474f2cf2.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "823c66c4d2568755e3f82a2a4f548d4a6c0e759e860941a018d201effa5a7241", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f703d40e709042f750aceb691c9fcac429d7e976e5601436019d74dc22799a33", "generation_seconds_batch": 18.187444249051623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.292664527893066, "genuineness": 3.848362922668457, "identity": 0.6202189922332764, "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_WARM_reg_high/823c66c4d2568755e3f82a2a4f548d4a6c0e759e860941a018d201effa5a7241.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "826a02cffa09b05268e2319a90ad30261369d8123490524fbda39dc94a5a3ffd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2d4fc48d07e0163dd3d38f407c7133c29ee2805f9c0becc1375422fd2dda0bcf", "generation_seconds_batch": 23.64911935303826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.149696350097656, "genuineness": 4.293961524963379, "identity": 0.6907058954238892, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/826a02cffa09b05268e2319a90ad30261369d8123490524fbda39dc94a5a3ffd.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "82e82251c142adfba3f16e36548d624f1b3add37e2e73ed9ccb1210ce0020498", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2a21fc6c2387ddede900b18c536778fef99dc58a3085663a89536185a5ad4ce5", "generation_seconds_batch": 18.08837638096884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6727256774902344, "target_z": 1.179687376489894, "ce": 6.086154937744141, "genuineness": 4.293961524963379, "identity": 0.6253541707992554, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/82e82251c142adfba3f16e36548d624f1b3add37e2e73ed9ccb1210ce0020498.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "830ad0d7eb74a4fc6c9da577c74827c78aa86d1a218f03c0d11203e6cf921761", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb3008b611d2815257bb892a0f7824f536a9da0eaff1040b2dbec73ab00cd388", "generation_seconds_batch": 17.74324284994509, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.5511944890022278, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/830ad0d7eb74a4fc6c9da577c74827c78aa86d1a218f03c0d11203e6cf921761.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "83406ab269c44a8555466f3d7408c1b39bbbbd570e6f5cb4251c71c5ce8859c1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "89f22d55338db5cf61b60733fa8d708346b0a0fa31371566a7283c293bb23ef7", "generation_seconds_batch": 17.667729720007628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.332377910614014, "genuineness": 3.898571252822876, "identity": 0.3187917172908783, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/83406ab269c44a8555466f3d7408c1b39bbbbd570e6f5cb4251c71c5ce8859c1.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "834c2e53a3121051b0db602796f9e9f724a90319893787a9adaea787a969918e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4e1de25d0bbf82184eeabee830d02ee48c119bd3aa4e017273ab6c855e539262", "generation_seconds_batch": 23.151764514972456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.252951145172119, "genuineness": 3.7918784618377686, "identity": 0.2193688154220581, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/834c2e53a3121051b0db602796f9e9f724a90319893787a9adaea787a969918e.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "836c254a722ddd3560aae7bfa3b983fae11bb8028395b909c699088703822b73", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ab34847567ee0a98d2c1faad0129576f3b6a3d4ed26f7327d063b7168a48ac2b", "generation_seconds_batch": 17.029371464974247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4076316356658936, "target_z": 0.8867187705850177, "ce": 6.221180438995361, "genuineness": 4.331617832183838, "identity": 0.7837066054344177, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/836c254a722ddd3560aae7bfa3b983fae11bb8028395b909c699088703822b73.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "8375f1da9f74a69173efa8f81f2b599f4ac095710420a7cbc34e96071ec0840d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "bdf68319891006ce9ac870f0a271eab5b8294b0bf6edf9ad3b57a2dc863663a5", "generation_seconds_batch": 23.47902188799344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6303107738494873, "target_z": 1.1328125576380494, "ce": 6.062326908111572, "genuineness": 4.720732688903809, "identity": 0.7008147239685059, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8375f1da9f74a69173efa8f81f2b599f4ac095710420a7cbc34e96071ec0840d.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "838aa8206716dc6a7d4bc8ab2cc3d3b57e66b59db5f41caeb8b1b06ef8f213a2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b54b5ca9546d48a3f4fb89b30c2bf044eb0b6633695374f1e00b48e0252e899", "generation_seconds_batch": 23.649388913996518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.358436107635498, "target_z": 1.9375000658720565, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.785602331161499, "hypothesis": "La porte 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_WARM_reg_high/838aa8206716dc6a7d4bc8ab2cc3d3b57e66b59db5f41caeb8b1b06ef8f213a2.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "839b2067c03be23df4bce01a1a46e19784183f2d0640d18188bd9e9fe916e209", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "231fe39924cb3167622a30ea8adab2d77df23ee5c2d6a9369d7578e48575f9b8", "generation_seconds_batch": 16.13402673904784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.324435234069824, "genuineness": 4.287685871124268, "identity": 0.3861325979232788, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/839b2067c03be23df4bce01a1a46e19784183f2d0640d18188bd9e9fe916e209.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "8465ae369af33e5fcc373b8088fa2bc335fffa4c3da09de23efde4aab55a836c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "90f77b100c3cfe4805349538dad8c73ca6d0e2ccef5cefcdc0de250b6a2fc4fd", "generation_seconds_batch": 23.27456595201511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.542234420776367, "target_z": 2.14062472004376, "ce": 5.569881439208984, "genuineness": 3.8546390533447266, "identity": 0.5513468980789185, "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_WARM_reg_high/8465ae369af33e5fcc373b8088fa2bc335fffa4c3da09de23efde4aab55a836c.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "848adb7927b28943ed76610ef005cb9b719765b26ecdcc0ea7df378a9c935a83", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4f3c0049dcc10d33c75f89560f12602ec23f7bbbd3562151f78090e45e314f15", "generation_seconds_batch": 17.569233587011695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.078212261199951, "genuineness": 4.243753433227539, "identity": 0.6636196374893188, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/848adb7927b28943ed76610ef005cb9b719765b26ecdcc0ea7df378a9c935a83.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "84d7b60a238f799c39e292e021bbac11528f2b52fc528955952d51d5d3e0afec", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e02c0e97f2c4a376fa59a07c3891fa35f1aa0aab7e86ad18588065848e6f14f6", "generation_seconds_batch": 17.534848776995204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.443574905395508, "genuineness": 4.15588903427124, "identity": 0.5510984659194946, "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_WARM_reg_high/84d7b60a238f799c39e292e021bbac11528f2b52fc528955952d51d5d3e0afec.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "84e16e5f949acb378428449efc54955668c723bc24d0632560aad00482721620", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "aa201acd3c30e28be26ae5f669952106b540208c42f752225caf1788da8e90cd", "generation_seconds_batch": 24.939703290932812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.173524379730225, "genuineness": 3.7165660858154297, "identity": 0.33057600259780884, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/84e16e5f949acb378428449efc54955668c723bc24d0632560aad00482721620.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "84fed9714573a4db2b9233eb3f436391f33495f818e6a95f1b61f1fd1a488ef4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a01114a507a434d2c28d796ef35fba29688d7f246eb112cab1a5b1901c735760", "generation_seconds_batch": 18.08837638096884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 6.324435234069824, "genuineness": 4.3881025314331055, "identity": 0.6572204232215881, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/84fed9714573a4db2b9233eb3f436391f33495f818e6a95f1b61f1fd1a488ef4.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "8514758b1036d8a2d578b2d48ce941342f675681976a68603b8eb5db421201ad", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e9fa8411e88314a143ac2fbd9a7c59d6438e96aa520b611067af0a8bc7a3ec1d", "generation_seconds_batch": 24.42609712597914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174637317657471, "target_z": 1.7343748847239013, "ce": 5.788304805755615, "genuineness": 4.356721878051758, "identity": 0.8225381374359131, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'est ce qui est.", "wer": 0.21052631578947367, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8514758b1036d8a2d578b2d48ce941342f675681976a68603b8eb5db421201ad.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "8524c209996115ea7fad00dc11f13b2ad964f0071eeb3d964206be2555f9d6a7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "63985c1d53944e1dbb6230819e53b534b099b56152b50765e89118d2741cee92", "generation_seconds_batch": 22.03417051397264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.213237762451172, "genuineness": 4.18099308013916, "identity": 0.4891555607318878, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8524c209996115ea7fad00dc11f13b2ad964f0071eeb3d964206be2555f9d6a7.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "85902bd50164a7a7da27e059fdefc2af7a9bc679c55762d6c1d7f1d25e6d4087", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "468f58bcf4f5e6d425b8b7eb738b63ce73da89ae520b00fb1a34c793bd39c665", "generation_seconds_batch": 25.214339010999538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.125868320465088, "genuineness": 3.898571252822876, "identity": 0.4120715260505676, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/85902bd50164a7a7da27e059fdefc2af7a9bc679c55762d6c1d7f1d25e6d4087.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "85df5565de964e05c0a19fece41d48de0a2f2ba8cc60c46d7d9387f8f0ab5a59", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fe95d02c0af84052f8946c9749b036ac809d1d0f8164a086f8ef9abaf360ac57", "generation_seconds_batch": 24.813943092012778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2415058612823486, "target_z": 0.7031249176599295, "ce": 6.372091293334961, "genuineness": 4.105680465698242, "identity": 0.7550694346427917, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/85df5565de964e05c0a19fece41d48de0a2f2ba8cc60c46d7d9387f8f0ab5a59.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "868d0b8d32dfb2e5adbd4b449829f010db31aa54caef7003791d51b88fea4a94", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a347b296bb7902c8e8a97d314db6ac10597dac8292fbe8c56de68e2c105bcfec", "generation_seconds_batch": 22.370217808056623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.316492557525635, "genuineness": 4.293961524963379, "identity": 0.6100703477859497, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/868d0b8d32dfb2e5adbd4b449829f010db31aa54caef7003791d51b88fea4a94.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "86c6659bc2d81d503e031e32e0cbe10b307e3a1f17f0bff4f26fe94cc57056aa", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fb2098d18a7333f383abb24e91e99d1a721153f00dcba49ec43a5351d16a2d3e", "generation_seconds_batch": 23.666640614974312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 5.998785495758057, "genuineness": 3.613011360168457, "identity": 0.6633632183074951, "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_WARM_reg_high/86c6659bc2d81d503e031e32e0cbe10b307e3a1f17f0bff4f26fe94cc57056aa.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "86d66ec86f22289cd7527288137b309d7486b974be8be4095a557fe7822b5db6", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "87b7872d01fddb4967beff980cca6e46c17b6806243849de41f5384764ff7784", "generation_seconds_batch": 16.324127857922576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.481858015060425, "target_z": 0.9687500329360282, "ce": 6.23706579208374, "genuineness": 4.268857479095459, "identity": 0.5680555105209351, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/86d66ec86f22289cd7527288137b309d7486b974be8be4095a557fe7822b5db6.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "86da4e5a488224fdbc578fcdebdb165133bd4f78f5652b8616a3887aa29fc174", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f5612f223b0e196f947061be93d098134465f2a53a72f2c52895e2debfdd16f1", "generation_seconds_batch": 26.659380369936116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.459460258483887, "genuineness": 4.118232727050781, "identity": 0.42519611120224, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/86da4e5a488224fdbc578fcdebdb165133bd4f78f5652b8616a3887aa29fc174.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "04a2f81c8be3cf08b1871b416dec0277b0b0730adc5ebe2927b0ee22d26c33e1", "generation_seconds_batch": 16.7208095099777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.523001670837402, "genuineness": 4.356721878051758, "identity": 0.2838926315307617, "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_WARM_reg_high/86f33165c5348ab0d60f28f425cc78e067d1a82beaed7f5f0c2b9e20a8fe6121.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "873238a02a8f2c7c87c00117274a8e789065fa8b6d4df6c4e72f9f33f0ac8bca", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "12b77f3b4ab06eec487f32b73faa37f8fced377cd527182dcc9b9a35d1fed7b3", "generation_seconds_batch": 21.542026336072013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.038498878479004, "genuineness": 4.1998209953308105, "identity": 0.4594138264656067, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/873238a02a8f2c7c87c00117274a8e789065fa8b6d4df6c4e72f9f33f0ac8bca.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "87ec9d858da4a3be866a65992612fe80de7014bbc82a7fe7512c24ec1ec15191", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "35c93bb67f8e94f88f258f45385b0692858bd04d98cf3e165b970ba5565a8d65", "generation_seconds_batch": 16.310697073000483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 6.0146708488464355, "genuineness": 4.319066047668457, "identity": 0.7691494226455688, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/87ec9d858da4a3be866a65992612fe80de7014bbc82a7fe7512c24ec1ec15191.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "87edf1d7b179a343358a1eb8b4882f09ce5b6749641d3cfc9fc89dbfed2091d3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "09fade64d0b5631f4ea17a3ba990815263f73b7123c5d44bda149c42c6771666", "generation_seconds_batch": 23.508138348930515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4182353019714355, "target_z": 0.8984374094259224, "ce": 6.229123115539551, "genuineness": 4.011539936065674, "identity": 0.6944383978843689, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/87edf1d7b179a343358a1eb8b4882f09ce5b6749641d3cfc9fc89dbfed2091d3.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "8870d1dc4a458e02202e434132996d4bcd7a11929ca783c313b063ed728dcaf6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "dc540d94fb94eb6cca2eaacdeba4591b63a03349cc678e9e8d78ea548eeff2fc", "generation_seconds_batch": 22.57222345902119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.259467601776123, "target_z": 1.8281250494040424, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.6141666769981384, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in koppelen.", "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_WARM_reg_high/8870d1dc4a458e02202e434132996d4bcd7a11929ca783c313b063ed728dcaf6.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "88b7bb729f0dccb19012df900504cead318e76e7ed514542c00294b710d4c3bc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "92741ce9e49ddc91a8d4a607e89817a431ee401ff008ea07cbb02df339ee0df4", "generation_seconds_batch": 21.507822999963537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.117925643920898, "genuineness": 4.231201171875, "identity": 0.6734894514083862, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/88b7bb729f0dccb19012df900504cead318e76e7ed514542c00294b710d4c3bc.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "88fd170cfeaac84c7b7f76bf30dedc8d1bc83501887a5e75fcc85e5a8efa7fe0", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "877febc21de7c44f376794f4b3a80409b4e020e7e0f911ac3daf3284b8c9322a", "generation_seconds_batch": 17.776912480941974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3015940189361572, "target_z": 0.769531328223067, "ce": 5.967014789581299, "genuineness": 4.369274139404297, "identity": 0.5760878324508667, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/88fd170cfeaac84c7b7f76bf30dedc8d1bc83501887a5e75fcc85e5a8efa7fe0.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "890706228e71a7527258f185d7e1bb87b9bfec1dc831a5713e82263a1c47af13", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2bca179606474e283c25c1a9b65ef309bcaea031a0d9ea53ab2f1bbe72dc7ccc", "generation_seconds_batch": 23.205246609984897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 5.685050010681152, "genuineness": 3.6412534713745117, "identity": 0.7097743153572083, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/890706228e71a7527258f185d7e1bb87b9bfec1dc831a5713e82263a1c47af13.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "894cd5371879b3f33c5f5a7e9e9466d16d38861da81d3df6334695eb04cd6cdb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1ed9149ab532ee7145aa2ebb8b923c63892002846ccbc80a5e76ef6560bee44d", "generation_seconds_batch": 24.464731225045398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.419747352600098, "genuineness": 4.482243061065674, "identity": 0.2659521698951721, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/894cd5371879b3f33c5f5a7e9e9466d16d38861da81d3df6334695eb04cd6cdb.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "89601e4e5b8242793e1a72dab509ed514cbcf6c6d697d270282077afb86e81de", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5dbac1df381825c53be02a592ea557808bf6fe90cbaa1e845124926113525d32", "generation_seconds_batch": 17.60515574994497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.117925643920898, "genuineness": 4.005263805389404, "identity": 0.2867182493209839, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/89601e4e5b8242793e1a72dab509ed514cbcf6c6d697d270282077afb86e81de.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "898014d155f06a3ef197fa075f5c80d8eb2f1f921ea807c6fd402ee52c731069", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b114c3a79cf74236ffb29323a6324fe207586fb6f23e02233b50bd33d948e74b", "generation_seconds_batch": 22.280806713039055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.118083953857422, "target_z": 1.6718749505959576, "ce": 6.141753673553467, "genuineness": 3.9299514293670654, "identity": 0.7493388652801514, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/898014d155f06a3ef197fa075f5c80d8eb2f1f921ea807c6fd402ee52c731069.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "898718217b62e45d65ed9a626d50310247546abdcd5e44cfb26df94bc0f1d128", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "490b7c5f3bdffe2fce96fdecf42540a879f8ba47d69ebee16b4039cbd17f2b26", "generation_seconds_batch": 23.34719469002448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 5.788304805755615, "genuineness": 4.099404335021973, "identity": 0.6442595720291138, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/898718217b62e45d65ed9a626d50310247546abdcd5e44cfb26df94bc0f1d128.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "89a5a272d394facde73dab0e5107154dc5da93af54828bd21d355904950d3ccb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "262d0bb2fb2b5bd67fd9a6fe400eabae070bd6266011e88d54b5e76482eecfeb", "generation_seconds_batch": 16.174031116999686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 6.1894097328186035, "genuineness": 4.068024158477783, "identity": 0.6080241203308105, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/89a5a272d394facde73dab0e5107154dc5da93af54828bd21d355904950d3ccb.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "660f53b63d14df702e6f896573543f6a6c5fa6e9840852907cd393d43b8ce6c4", "generation_seconds_batch": 17.187940155970864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.202914237976074, "target_z": 1.7656251152760987, "ce": 6.252951145172119, "genuineness": 4.068024158477783, "identity": 0.7312709093093872, "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_WARM_reg_high/89e0024d19b0aba55264e5761ece9d60a183dd13825a017339f901aacf2cd363.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "8a19d1c5938fafda53211b61528c298f6c53edf06799785e69e824d3cf190fa8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "86088a564e82047935e6b54e13c3d676a202fbd7e762ce68aca3c3a69a8245c4", "generation_seconds_batch": 24.07497872703243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.7439557313919067, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8a19d1c5938fafda53211b61528c298f6c53edf06799785e69e824d3cf190fa8.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "8a57f1498cc93301bea483f5a9c1a56fa5c6cafa191c519d03741c8683a47bf0", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9986baef9f81afcdcc97cd3d27912d7ef2707587f5fa508724dbae41e4d206d8", "generation_seconds_batch": 17.469084804994054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.6504151225090027, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8a57f1498cc93301bea483f5a9c1a56fa5c6cafa191c519d03741c8683a47bf0.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "8aa6c428a90859b45b8dfe7484c1a7c12e4bdfd3f2cf13cb6c7eae3cd8ab89d4", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4ccc0b70394887767cd7d35e1eae24f81fea8a91255ca232cfd657b09fdb9e68", "generation_seconds_batch": 18.00894882308785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.443574905395508, "genuineness": 4.3127899169921875, "identity": 0.6271864771842957, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8aa6c428a90859b45b8dfe7484c1a7c12e4bdfd3f2cf13cb6c7eae3cd8ab89d4.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "8ad3b6fc48488fcfcfd83c9f0d33762166cc4ad967f15db7bec5e7a91f1cd270", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f158bafe29a84cec0115a269ef18153127aa075c045d71b21b43501cd99e5198", "generation_seconds_batch": 17.16057678195648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 5.649308204650879, "genuineness": 3.7165660858154297, "identity": 0.488309383392334, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8ad3b6fc48488fcfcfd83c9f0d33762166cc4ad967f15db7bec5e7a91f1cd270.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "8b37146a52a9a788c845e50dc2ff077b2bdf0936edd73660892dc696d457ee79", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "87af9e885838c5fd95c54b2eae6c2bda0da479ad832420f982e4d96e655cd8ed", "generation_seconds_batch": 17.37557801499497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.197352409362793, "genuineness": 4.250029563903809, "identity": 0.1776810735464096, "hypothesis": "I found the little box. It was beside the old window. We can look inside too.", "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_WARM_reg_high/8b37146a52a9a788c845e50dc2ff077b2bdf0936edd73660892dc696d457ee79.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "958dfdd832e6a0fe4ade8ca60899e68183768d60c6a4b753f722f5a68f932e3c", "generation_seconds_batch": 17.795576908043586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.38003396987915, "genuineness": 4.545003414154053, "identity": 0.6421116590499878, "hypothesis": "Ehemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "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_WARM_reg_high/8b597a76b05dbc4709eaacf724e876207d67a466eac6fdd1f8c08cfeaab41841.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "8bbabf105e17caf47aa348e81e59098d0b051496572bfb59b0aa64d50e4a2cd6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b23b5ea9d818ee55327b479f1dabb0ac436540693d12432878a9eb0abce5f296", "generation_seconds_batch": 23.435381450108252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8138206005096436, "target_z": 0.23046867177693298, "ce": 3.258563995361328, "genuineness": 3.6914620399475098, "identity": 0.30323493480682373, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8bbabf105e17caf47aa348e81e59098d0b051496572bfb59b0aa64d50e4a2cd6.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "8c157afb356316de0768562e00fa3746ae205d5c9f80a0fd40925c2ea216b8b1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "869bb3d3d89bf4c66cc3d57d41149329ee23e550a153317391c89d811f95414b", "generation_seconds_batch": 21.084210560889915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.308549880981445, "genuineness": 4.3881025314331055, "identity": 0.39006149768829346, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8c157afb356316de0768562e00fa3746ae205d5c9f80a0fd40925c2ea216b8b1.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "8c795185ca82ba34eea62fe42e85439b57037b421481d3e12d6fab989d1897c6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "322883c752095d99fa00d35a456013a09f83c723b5c348f8ad1dcf3f1cbe6cdd", "generation_seconds_batch": 17.764955307007767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.268836498260498, "genuineness": 4.162164688110352, "identity": 0.6736208200454712, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8c795185ca82ba34eea62fe42e85439b57037b421481d3e12d6fab989d1897c6.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "8c7f58af5d3a86f41070dbe36c54528e1eea080b1d2bf710726140559525f0ee", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "15d8d697c277c67901890f3d2b4ede28197f5912228894639d5658e22f5e2cb5", "generation_seconds_batch": 22.918663172982633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7151408195495605, "target_z": 1.2265624588299646, "ce": 6.078212261199951, "genuineness": 4.024092197418213, "identity": 0.7184223532676697, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8c7f58af5d3a86f41070dbe36c54528e1eea080b1d2bf710726140559525f0ee.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "8d455aeae7138b32cf81f86f5c8d565ec5915f0fd05db840afca5b759536f347", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a1192bc1e2f3c4cb62318ebfeabcb05e10ba9251a345ebdd6e23f8196b4d3b9a", "generation_seconds_batch": 21.91711525397841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.157639026641846, "genuineness": 3.8044304847717285, "identity": 0.6722827553749084, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8d455aeae7138b32cf81f86f5c8d565ec5915f0fd05db840afca5b759536f347.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "8d762a1414182b8f624e8de8d891fcbd17d6db30cf7c268093bf80e1b524d4da", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b392a78167b51621312b74e0a45ddec3020093fc3c17918ca3244bc507aead3d", "generation_seconds_batch": 23.44479463400785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.5080447196960449, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8d762a1414182b8f624e8de8d891fcbd17d6db30cf7c268093bf80e1b524d4da.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "867fed37d05659e15881243c47639ce26b0011ca23cf690145296fea47a26612", "generation_seconds_batch": 17.50107441598084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089807510375977, "target_z": 1.6406252470202116, "ce": 6.3641486167907715, "genuineness": 4.2249250411987305, "identity": 0.4059944450855255, "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_WARM_reg_high/8d7e7b4d146149fd4d853dd6ee74feb094ce1866344c602c7e06aac880da05a6.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "8d9a34cc8d32ddb45b8ee4ad37fb935fef2c8325ed4d3b37e0cf1760b3b0edb9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8e75a250f8f942694f1332a4d3bead028b4dae67c1e2136bb37bdfe334ac01ca", "generation_seconds_batch": 16.106708309962414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.259467601776123, "target_z": 1.8281250494040424, "ce": 6.4515180587768555, "genuineness": 4.331617832183838, "identity": 0.6768636703491211, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8d9a34cc8d32ddb45b8ee4ad37fb935fef2c8325ed4d3b37e0cf1760b3b0edb9.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "8e164288c5d7ed1025598fe6fabfa4eadead12c6ed4092c92b77bbf55a939d15", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f498f0e003c4a15b928aaf712a07b997dec7566e542ebb71647e3780f9285696", "generation_seconds_batch": 23.439149915007874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457404613494873, "target_z": 2.0468750823400708, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.7071859836578369, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "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_WARM_reg_high/8e164288c5d7ed1025598fe6fabfa4eadead12c6ed4092c92b77bbf55a939d15.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "8eb8137b63700243e729d029f4ef490db3128ec301fbb2721be20494d7376f2f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "754b8d9b73fe7a24db0135cf2e8a5e34c2c04a2e6a8fed48d736a60227de3e9d", "generation_seconds_batch": 25.700152426958084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2099833488464355, "target_z": 1.7734375411700354, "ce": 6.252951145172119, "genuineness": 4.149612903594971, "identity": 0.3504042625427246, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8eb8137b63700243e729d029f4ef490db3128ec301fbb2721be20494d7376f2f.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "8f6cb17703d4399ca8f34227649212f9dece24575fa0eb151d465b4006cac92e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f0b955926fdf83936605f5db6db1b46917940fa5887f62657342e7289222bcf4", "generation_seconds_batch": 17.78092242591083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174637317657471, "target_z": 1.7343748847239013, "ce": 6.022613525390625, "genuineness": 3.9927117824554443, "identity": 0.49932536482810974, "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_WARM_reg_high/8f6cb17703d4399ca8f34227649212f9dece24575fa0eb151d465b4006cac92e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "8fb207c03a4d59ab9d00345784929d3d0c3e077c6fa34557a07bb8e4bb7320d6", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "07cb6b96f12a37231bcae1b31f0c0fd3269c7ab504d02e28fdaecfe270c3ca34", "generation_seconds_batch": 17.18539562600199, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4394428730010986, "target_z": 0.9218749505959577, "ce": 6.086154937744141, "genuineness": 4.049196243286133, "identity": 0.8294959664344788, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8fb207c03a4d59ab9d00345784929d3d0c3e077c6fa34557a07bb8e4bb7320d6.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "8fb9a7281a4b7aaba7341db84920ccb674a3822df44a7f1feee86f2406daac0b", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c4b4fd50a405d59f50078fe5e6d68661831a516761dc2b43db02fde582f376d6", "generation_seconds_batch": 16.81424726091791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.481858015060425, "target_z": 0.9687500329360282, "ce": 6.284721851348877, "genuineness": 4.444586753845215, "identity": 0.45473015308380127, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/8fb9a7281a4b7aaba7341db84920ccb674a3822df44a7f1feee86f2406daac0b.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "9012609faabf118017c53816c5b6b03141671d593e2cd2742463ccb82f8c85b3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0448b5c557425e1d892648ebe6e1344faa14a06804fdefa41dd416cd72a9a01a", "generation_seconds_batch": 21.957208935986273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.598788261413574, "target_z": 2.203125181148155, "ce": 5.982900142669678, "genuineness": 4.394378662109375, "identity": 0.688843846321106, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9012609faabf118017c53816c5b6b03141671d593e2cd2742463ccb82f8c85b3.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "903171bdde96b9e7520d108d51b9ae95170306d942fe352cb2ab75f28355d573", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "26785ff3491f28688f6c7b6ab7a3b9e996679e156672f087d8c5f4fcd8c4647b", "generation_seconds_batch": 17.21377755701542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.483288764953613, "genuineness": 4.306513786315918, "identity": 0.6024447083473206, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/903171bdde96b9e7520d108d51b9ae95170306d942fe352cb2ab75f28355d573.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "90c46567e2aae8d1b327f05460678f33cfe2295bf4b131943e6441a2fbc9e72a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d98dd9a7cf58ab0fed4565945c49ccb62bcaa11d7fbb1337bf822f330e8c5a1a", "generation_seconds_batch": 20.86953905900009, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.4912309646606445, "genuineness": 4.406930446624756, "identity": 0.5346204042434692, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/90c46567e2aae8d1b327f05460678f33cfe2295bf4b131943e6441a2fbc9e72a.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "90f953eafffdf0dcfdccfc42a0fb5d739bae9f1e9221a2ca626cd95a99f9c2b9", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7acccf1517fbff1e294de7c3862a3a60de2a0eb61a1b53714cb1616f4a9fe8e1", "generation_seconds_batch": 18.075655404943973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.213237762451172, "genuineness": 4.331617832183838, "identity": 0.774056613445282, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/90f953eafffdf0dcfdccfc42a0fb5d739bae9f1e9221a2ca626cd95a99f9c2b9.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "918a13d746bc8b945dc9e568bd5cce05b17ec6e516494bbb3149ad46dd390fd2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4467b3682ed3a74d25be1929d69fef4199b119476080c3c2fc4045580ac55f18", "generation_seconds_batch": 25.15820707299281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.24500846862793, "genuineness": 4.15588903427124, "identity": 0.4456750750541687, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/918a13d746bc8b945dc9e568bd5cce05b17ec6e516494bbb3149ad46dd390fd2.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "91bcc1280987e5b45f46e6923a50c3f179910000f59e927d8f0db2e336f1022f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "93b6f4ab6abef88f9ee0c3de44d834d3e37246134e9de98382f30fbeaeb855c4", "generation_seconds_batch": 15.946987231029198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.332377910614014, "genuineness": 4.243753433227539, "identity": 0.3517240881919861, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/91bcc1280987e5b45f46e6923a50c3f179910000f59e927d8f0db2e336f1022f.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "921a20154499271448e652f8766feda95714e4d13b8ecf21e2a0d94b2a701ca6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "981c68c05ee66b765b3c315f2092fca856c89a890cd6048005764b38cd79c1fb", "generation_seconds_batch": 22.393970516044647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.7053004503250122, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/921a20154499271448e652f8766feda95714e4d13b8ecf21e2a0d94b2a701ca6.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "9240c7e88030421d8497dc147cd0c3d619150a5c1f3e1720b5e775cf168ab9a3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0c17d66ffb9eba73d091515dc393d16cbb0c46c6133917c1d9c79cd4b6155b59", "generation_seconds_batch": 23.875661403057165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.260893821716309, "genuineness": 3.8797430992126465, "identity": 0.803776204586029, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9240c7e88030421d8497dc147cd0c3d619150a5c1f3e1720b5e775cf168ab9a3.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "9245969059124abd9eb8b9bf6aa434e209946e9c8a6fad53f05e10be581ead2e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "59819655ae216d40c8b5eb98aa28f6a86887b88110ec84a7518a17b8f1954ce3", "generation_seconds_batch": 23.00109107105527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.221180438995361, "genuineness": 4.482243061065674, "identity": 0.579442024230957, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9245969059124abd9eb8b9bf6aa434e209946e9c8a6fad53f05e10be581ead2e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "924e697cdc6649275f06d96fb9d20645f2a03b28b5eea4b5edc62b2f2f0c3780", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3137afefb8239e283cc967a3200eeca9cdc24e45082e3b25facfe6666f55002d", "generation_seconds_batch": 17.18539562600199, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.252951145172119, "genuineness": 4.444586753845215, "identity": 0.7909665703773499, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/924e697cdc6649275f06d96fb9d20645f2a03b28b5eea4b5edc62b2f2f0c3780.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "9273cd4171cf9d2d4e81eceebee282f284916fce2c37d796e589e57f84b6034d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4d1ad83324c5dd3d4812555cbaeb374236871f2ad49cd488fc04a830ef55121c", "generation_seconds_batch": 21.59061174991075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.141753673553467, "genuineness": 4.030367851257324, "identity": 0.7507767677307129, "hypothesis": "De deuren, de deuren, de deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.21052631578947367, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9273cd4171cf9d2d4e81eceebee282f284916fce2c37d796e589e57f84b6034d.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "92a29283a83633c295f7c59be72251701bd0b4247c0e4c926770b22304c0ecef", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4dd04a9c19cb452caa3f577f7c3c3d109183c5cbd54d91c0c4859b264e634725", "generation_seconds_batch": 22.17250768793747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.8150674700737, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/92a29283a83633c295f7c59be72251701bd0b4247c0e4c926770b22304c0ecef.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4529cb85e195958953810b15f90acbeac281bac07135553061ebd8c9644bbaa6", "generation_seconds_batch": 21.225058619049378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.828247547149658, "target_z": 1.3515623270858519, "ce": 6.252951145172119, "genuineness": 4.344170093536377, "identity": 0.7560440897941589, "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_WARM_reg_high/92b1bcadb18d93f3c0a0d9a8645d936b17a3b1a8ad73abc25dbbb2f2dc7ec60b.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "92b32c1acb7755feda3f46e1b72bfbd633ca662d556db6cb970b1db2650462fc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2c195edf5a52b7c6ebd1d3619eb36e0d00117e828944b05859b859c4bccd4899", "generation_seconds_batch": 21.756471942993812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.205295085906982, "genuineness": 4.300237655639648, "identity": 0.5338538885116577, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/92b32c1acb7755feda3f46e1b72bfbd633ca662d556db6cb970b1db2650462fc.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "92c0563b6146f325bb69d5bf90d0f2bb578170c39e3d367588ec33ec6daf2663", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "108281c0f62812c53b47e87f14cc7a31a0a341a6538b7b9f367a665845707280", "generation_seconds_batch": 22.09716922591906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.153429985046387, "target_z": 1.7109376070420916, "ce": 6.006728172302246, "genuineness": 4.482243061065674, "identity": 0.786203920841217, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/92c0563b6146f325bb69d5bf90d0f2bb578170c39e3d367588ec33ec6daf2663.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "92dacb255b7182484e2259277855fe6d60344f92d506b66f0b0207facf60a669", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8703988f8eb4706985945df09ac099f5f0f9f864e2e6116dd44fa590976e9dbf", "generation_seconds_batch": 17.57428386399988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.221180438995361, "genuineness": 4.13078498840332, "identity": 0.5138385891914368, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/92dacb255b7182484e2259277855fe6d60344f92d506b66f0b0207facf60a669.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "933d947a1b0241796758ac4b288b650cba84ee24c29008a2411986da79b2fb8e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e5691720d7b79aca2d271e10ae0d20a5caad180bac607df33f87677cb33f052d", "generation_seconds_batch": 23.613387380028144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.6670206189155579, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/933d947a1b0241796758ac4b288b650cba84ee24c29008a2411986da79b2fb8e.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "948f9c4cc9e9a186e5f84756f8be7d9ae97ee5176d2357cb5ef0278e8b1b0d30", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "0b26b264e8ea6f7d551b4648b097b75aab017df4a531bd1aeda76353699b4b8e", "generation_seconds_batch": 17.16057678195648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.259467601776123, "target_z": 1.8281250494040424, "ce": 5.673136234283447, "genuineness": 3.7542223930358887, "identity": 0.6150091886520386, "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_WARM_reg_high/948f9c4cc9e9a186e5f84756f8be7d9ae97ee5176d2357cb5ef0278e8b1b0d30.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "949815969207f2097847f71be7861edd58dd8b34fcf2ff9d2c910f2cbc4b9b91", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ea375ad5e42eeddaad3d617624856812af1b9bdf35045879dde946dcbc8cba42", "generation_seconds_batch": 23.000958855031058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.488927125930786, "target_z": 0.9765624588299647, "ce": 6.125868320465088, "genuineness": 4.15588903427124, "identity": 0.7318230271339417, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/949815969207f2097847f71be7861edd58dd8b34fcf2ff9d2c910f2cbc4b9b91.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "301ccd9197043760fe75eac078710622130393a85af20cac0cb0a817435ccb0d", "generation_seconds_batch": 20.426329857087694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.133810997009277, "genuineness": 4.444586753845215, "identity": 0.775137186050415, "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_WARM_reg_high/94b51ea9823373604af8d99886d2c720c2969c7e9f47f866df396069127228f2.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "94c52a50cd3fb8ca83c0ea99004b8273c63e2497bb0237762b0baa2b8f3968b9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4a24b184f8b9c181d3ba3d1f6dcd367a4deff58e0fd26e60ffb1a409d78adf98", "generation_seconds_batch": 23.959351639961824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.197352409362793, "genuineness": 4.118232727050781, "identity": 0.6830266714096069, "hypothesis": "Ik 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_WARM_reg_high/94c52a50cd3fb8ca83c0ea99004b8273c63e2497bb0237762b0baa2b8f3968b9.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "95533e3cfc72ac363ec7ddec7b59ded8cac6f8d98e9cb953efcdc5c55412c9e7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "24c142d415e8c7ab9f89ca8576acf4c8f79c33ee91757e68872b56c359d08d25", "generation_seconds_batch": 25.480022610980086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.340320587158203, "genuineness": 4.174716949462891, "identity": 0.749269425868988, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/95533e3cfc72ac363ec7ddec7b59ded8cac6f8d98e9cb953efcdc5c55412c9e7.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "95665f5d62a92ad8ce24c14394ec3eb48013d8a8262e496c9b61142ad9d3d942", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "0dc925fad6000583babed794a1803c5b320292504e15c7386bbef86596bf1cfc", "generation_seconds_batch": 18.50474062305875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.252951145172119, "genuineness": 4.425758361816406, "identity": 0.7004727125167847, "hypothesis": "Deine Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/95665f5d62a92ad8ce24c14394ec3eb48013d8a8262e496c9b61142ad9d3d942.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "958d87ea9e71689c6b73e4f5e3fe1c776793c8618e248eccc06ec71bdc4fe7bb", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "89370e5dd53d9250ab9b96a35a6c0a80a2c20ea7e2ed526ad11201aad611c299", "generation_seconds_batch": 15.87749464600347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.181467056274414, "genuineness": 4.545003414154053, "identity": 0.7474919557571411, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/958d87ea9e71689c6b73e4f5e3fe1c776793c8618e248eccc06ec71bdc4fe7bb.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "95a486e28d220dbf8268c46a870977da90124946c7af1ffcd30bc9cbc9c6ef58", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d0141186b33ec38b229ac959cf45a4398a6a59c70253e8ad5adf0ca9a3378768", "generation_seconds_batch": 23.184207169921137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5172038078308105, "target_z": 1.0078124258939365, "ce": 5.998785495758057, "genuineness": 3.9425034523010254, "identity": 0.5604792833328247, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/95a486e28d220dbf8268c46a870977da90124946c7af1ffcd30bc9cbc9c6ef58.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "967340e3b18d94c0b7a35305177daa2c1a26fd6b8d6d50dcedc8fa6e51e2df20", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "59ae67ac75f4207b39afa5cf7fd853930c913d44c28ca0422c9f2979ef9ce924", "generation_seconds_batch": 24.542143414961174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6727256774902344, "target_z": 1.179687376489894, "ce": 6.483288764953613, "genuineness": 4.494794845581055, "identity": 0.5511680841445923, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/967340e3b18d94c0b7a35305177daa2c1a26fd6b8d6d50dcedc8fa6e51e2df20.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "96d764f89dfb9179288331dacb031cdc84714ac7bc8de9abb7711c44e7b69441", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "380bd441521a664eab14f0861a9f9d7b379a6dec17754136d00746f23c472697", "generation_seconds_batch": 22.47985800402239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.38288950920105, "target_z": 0.8593750164680141, "ce": 6.149696350097656, "genuineness": 3.948779582977295, "identity": 0.34038060903549194, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/96d764f89dfb9179288331dacb031cdc84714ac7bc8de9abb7711c44e7b69441.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "96deef5f8f95b673b0b2be9347865accfafe69a35a6a09b9664c7bf6e1cc1fe8", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e4fcdf33d56f8a8eea646e57a6bcdb11f3a37364c6e3f715e5fa9914b4d6951a", "generation_seconds_batch": 15.663766894955188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2768518924713135, "target_z": 0.7421875741060635, "ce": 6.062326908111572, "genuineness": 4.243753433227539, "identity": 0.665682315826416, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/96deef5f8f95b673b0b2be9347865accfafe69a35a6a09b9664c7bf6e1cc1fe8.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "0b4419ab79aef9ce8bd39ad8e253d7b0d9813d85dbc3ae0a280e2fadc39b5213", "generation_seconds_batch": 20.40659520495683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.139291763305664, "target_z": 1.6953127552542187, "ce": 5.998785495758057, "genuineness": 4.482243061065674, "identity": 0.6839211583137512, "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_WARM_reg_high/97921a799c4e8bf9c1cd60b11357cc91d07ad263ba54ca307924efa2fd17e560.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "9801b1e462440673ddb2103af13872ab5dc67f2b9acd787b49fa9496aa87ba7e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4e1314e50b5460c5fe43ab0848edfd46bb31cf1b904648e1fab0e7c34e8f1e82", "generation_seconds_batch": 23.534055243013427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 5.824047088623047, "genuineness": 4.005263805389404, "identity": 0.48552095890045166, "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_WARM_reg_high/9801b1e462440673ddb2103af13872ab5dc67f2b9acd787b49fa9496aa87ba7e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "982ef947c9c13a85df933f88b407d572fbac93c33f394d50d2a2d11683efdc72", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a1f4672ebec1a6a3fcf5f7b2ee0c218e3225a342d7c9e4716846af7c8fa50089", "generation_seconds_batch": 24.422171998070553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8141093254089355, "target_z": 1.335937475297979, "ce": 6.0146708488464355, "genuineness": 4.833701133728027, "identity": 0.6261054873466492, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/982ef947c9c13a85df933f88b407d572fbac93c33f394d50d2a2d11683efdc72.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "985d8d25256e7dbbb1d192d712dca263e70b02bbbde2350624d711bc18fb008d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ba04c26bf16bf275fb3205597e4869984c041c915f2de477a87c263daaa45427", "generation_seconds_batch": 17.199293129961006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 6.268836498260498, "genuineness": 4.482243061065674, "identity": 0.726087749004364, "hypothesis": "De deur 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_WARM_reg_high/985d8d25256e7dbbb1d192d712dca263e70b02bbbde2350624d711bc18fb008d.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "985e8ca7aa9b0baca147a70ed7cbf5e28af14ccc36507835b352475a3e7c6f08", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "64cfa87d7af2a9cf5ab49f626a49581c33108d934f88e4b959d82449c054f0f8", "generation_seconds_batch": 22.113278625998646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.167568206787109, "target_z": 1.7265624588299646, "ce": 6.125868320465088, "genuineness": 4.268857479095459, "identity": 0.8132691383361816, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/985e8ca7aa9b0baca147a70ed7cbf5e28af14ccc36507835b352475a3e7c6f08.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "98d6e0c1ebed7d45ddad8fa9b9ebc4356a93aa20f4feeec58fdf523d59fa8b80", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6b41fc1b39f9cce62ec291cf056dbd0db12a547b67ab221269a7c76f40eb7514", "generation_seconds_batch": 22.72654182394035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 5.903473377227783, "genuineness": 4.231201171875, "identity": 0.6621613502502441, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/98d6e0c1ebed7d45ddad8fa9b9ebc4356a93aa20f4feeec58fdf523d59fa8b80.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f60774ebc4e903012cab561b3705507b1a322791662bf19c8c74e6c259d75cd6", "generation_seconds_batch": 33.26402025995776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.356205940246582, "genuineness": 4.143336772918701, "identity": 0.1810699701309204, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/993ca0359509b049d32ca8ecb6c24a89a4eb20c2ca5325d25284930db2d5722b.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "9952025383fc96782fabc2d70f8fb5d12c425f0a81366a40cf0d386fb6fa5e73", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b4dda8335be80d0f07074704395de97b38990edf7116223103827cc2fd23137c", "generation_seconds_batch": 23.37862551410217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.165581703186035, "genuineness": 4.381826400756836, "identity": 0.7066069841384888, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9952025383fc96782fabc2d70f8fb5d12c425f0a81366a40cf0d386fb6fa5e73.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(very warm vocal delivery. 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": "7b4b06f3d5b61a42f8d6d8153ff0bf580c0a28b2f63a81d155a7cbff4d6cca27", "generation_seconds_batch": 15.431697850930505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.308549880981445, "genuineness": 4.570107460021973, "identity": 0.6501361727714539, "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_WARM_reg_high/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "99dfc91321ff7a6be3fd6aa892283cbb1268b2888266f4ce28ddda2f9f4e2897", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b74eba4d04b73c697d3fdb24f16cbd6fd8b71838fcaf8f915fa365d3661aa417", "generation_seconds_batch": 17.69070610392373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.0146708488464355, "genuineness": 4.720732688903809, "identity": 0.6172164678573608, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/99dfc91321ff7a6be3fd6aa892283cbb1268b2888266f4ce28ddda2f9f4e2897.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "9a0b66effbf87270dbf420c2fb0a16ab8fdb39a357bda9e072de955949aa6ce8", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2f2966a36988db9f63f82071074f06ea277cd1d759fb24753a2cedeae74bc6bd", "generation_seconds_batch": 16.347157882992178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.149696350097656, "genuineness": 4.645420074462891, "identity": 0.7220990061759949, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9a0b66effbf87270dbf420c2fb0a16ab8fdb39a357bda9e072de955949aa6ce8.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "eb5467bcfc4cd81c7b4aad0575481c10dffc6be473cc5e867cd726a765aa6d43", "generation_seconds_batch": 17.144357090000995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.348263263702393, "genuineness": 4.011539936065674, "identity": 0.4093511998653412, "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_WARM_reg_high/9a289004acaf14570fba6ebfecdd244d6362c3ebfd60e0fa8f9410f1c1762e38.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "9a32031fb277264ee40ddf952243bd1fb924a299b79994d3d94b154c3db74f34", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "bfb85073285fd3ac296527e4a2cf08d2c1e843c95288fa63e770aad0e58f3fcf", "generation_seconds_batch": 17.64097264909651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6727256774902344, "target_z": 1.179687376489894, "ce": 6.125868320465088, "genuineness": 4.243753433227539, "identity": 0.5515934228897095, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9a32031fb277264ee40ddf952243bd1fb924a299b79994d3d94b154c3db74f34.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "9a63a342b55d7be93c03c5e6df6a531880d4e126029aac30683d28aafa32dfad", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3b55263379db62baec02d7e94a33ed1720d5d4f28c785f6380a05b71bbafa506", "generation_seconds_batch": 24.519808424985968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0120463371276855, "target_z": 1.554687508234007, "ce": 6.213237762451172, "genuineness": 4.632867813110352, "identity": 0.5386090278625488, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9a63a342b55d7be93c03c5e6df6a531880d4e126029aac30683d28aafa32dfad.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "9ac4708a412834711a0946fe6cd229f309adb035f3ddb313acd7a1dc112eb8a0", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f8e4edb915b52665c5be6d96941367db31960bda1b2567252e2db1d66c9e8601", "generation_seconds_batch": 17.52722038794309, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 6.181467056274414, "genuineness": 4.118232727050781, "identity": 0.6703716516494751, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9ac4708a412834711a0946fe6cd229f309adb035f3ddb313acd7a1dc112eb8a0.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "9b007dfd3b5d6c4a15ba576172190cd044c244cbc6bf31b2a8e1de99448eb127", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e456bd764a72ff8c1c95653ff6c22b2621a43c09abc2a86bedf3ddc85e30f3a1", "generation_seconds_batch": 25.469032046967186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.326336145401001, "target_z": 0.7968750823400705, "ce": 6.41180419921875, "genuineness": 4.557555198669434, "identity": 0.32612916827201843, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9b007dfd3b5d6c4a15ba576172190cd044c244cbc6bf31b2a8e1de99448eb127.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "9b5a5fc7b8fa225cc1f390a096dbf5d3467a0d7202983239651efa314bd13e47", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ebc82168e893bcfa79f6b5475d1662fd70cb6fa2f63dd825a1223290b25216d3", "generation_seconds_batch": 23.44711526297033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.197352409362793, "genuineness": 4.193545341491699, "identity": 0.6342349052429199, "hypothesis": "Die Tür 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_WARM_reg_high/9b5a5fc7b8fa225cc1f390a096dbf5d3467a0d7202983239651efa314bd13e47.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "9bebb7241ca8900c87abe50da631e16850f6b4d378cb6da2c687a84cdf3461b9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "05a41695d6266833831229fc79962d17e588c54bbd6f212a5cf548dc0303b390", "generation_seconds_batch": 28.0569482780993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.054384231567383, "genuineness": 4.432034492492676, "identity": 0.8324713706970215, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9bebb7241ca8900c87abe50da631e16850f6b4d378cb6da2c687a84cdf3461b9.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "30c329baf4e163824b0635db054c46b459da074449d602f9de3bac0701ce43ed", "generation_seconds_batch": 21.156972769997083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.6709315776824951, "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_WARM_reg_high/9bf46ed83e2e31a78e17d8f9f74a2663a9672c097d86c00eaa568cdf8fb4a1e3.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "9c030268dfb25ba196fcc55924cc1fa21a5b10a7ae0c7d280c3b0427444d0f50", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "145b5b1634bac3c7e7807ff3a9c6c56685d9e33f6588016c2445886c531d49f5", "generation_seconds_batch": 22.816591658978723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.443574905395508, "genuineness": 4.444586753845215, "identity": 0.5041826963424683, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9c030268dfb25ba196fcc55924cc1fa21a5b10a7ae0c7d280c3b0427444d0f50.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "9c05ac61fd26eab83be415c158604c99c1400664238a137dd5b14df686cc7918", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "658ff8178ca9f295174877febebd731dae35ec96adf5ed6af50625249c30b7d9", "generation_seconds_batch": 23.17728814308066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.395919322967529, "genuineness": 4.143336772918701, "identity": 0.567146360874176, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9c05ac61fd26eab83be415c158604c99c1400664238a137dd5b14df686cc7918.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very warm vocal delivery. 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": "17c77ae9b65c035893f00cd4e760595a97447b8013700ff75e693d2123325973", "generation_seconds_batch": 16.79091814206913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5066001415252686, "target_z": 0.9960937870530318, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.7533178329467773, "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_WARM_reg_high/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3", "bucket": "vn_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very warm vocal delivery. 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": "8baa487947fc841d990404882c8ea6607d0ccdc739d1f0308cce3e70de70e51f", "generation_seconds_batch": 16.301049164962023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 6.340320587158203, "genuineness": 4.6203155517578125, "identity": 0.7461427450180054, "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_WARM_reg_high/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "9d49332e904bf0a04b813db45a191dec6052da26b57e304a601d0807f72416a3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8dd3ffd8fa8167358360822fafe67c0e09855f0cbcb4ac34368a106a06343945", "generation_seconds_batch": 23.03621890710201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.745396614074707, "genuineness": 4.482243061065674, "identity": 0.5977500677108765, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9d49332e904bf0a04b813db45a191dec6052da26b57e304a601d0807f72416a3.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "9d738f2251f4faa720f8a4c698f0c4397db97c48e976e8636fa1861d00bfd422", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "039a4c3c39b8634dfcd02fc12961d9d94d2076513071c4484750a31a62e1d408", "generation_seconds_batch": 17.029371464974247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.324435234069824, "genuineness": 3.860914945602417, "identity": 0.5401390790939331, "hypothesis": "Ik 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_WARM_reg_high/9d738f2251f4faa720f8a4c698f0c4397db97c48e976e8636fa1861d00bfd422.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "9e231fb67b5339dd1c9608767a022df78dc0503401fbf14bf7f131e7d3f585d3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c113a804446e13ac22c3de27d7693a6fecc2ce96f454f93f49491a35afe31029", "generation_seconds_batch": 23.34719469002448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344297409057617, "target_z": 1.921874687107732, "ce": 6.070269584655762, "genuineness": 4.18726921081543, "identity": 0.3853365182876587, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je.", "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_WARM_reg_high/9e231fb67b5339dd1c9608767a022df78dc0503401fbf14bf7f131e7d3f585d3.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "9e69c1866bee49b87c843f7de2d7a4c5f86a6e7d7c5d4a2b3fda1cde254e9759", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "41a61d2055608ef901e85ac0e5e3dc02b6c381a013284927de6d62c97374fd56", "generation_seconds_batch": 23.85519957402721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.273605823516846, "target_z": 1.8437499011919154, "ce": 6.395919322967529, "genuineness": 4.281409740447998, "identity": 0.48826465010643005, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9e69c1866bee49b87c843f7de2d7a4c5f86a6e7d7c5d4a2b3fda1cde254e9759.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e41d7b1693f5a21eb2201393d83a5e0413eba34cf6f21cf5b8c35f105a3f5e29", "generation_seconds_batch": 20.993672338081524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.372091293334961, "genuineness": 4.457139015197754, "identity": 0.7684306502342224, "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_WARM_reg_high/9e8cfc97467e7ee3ffa4187e2df1dd8b7e68c8481bf1c668c2c63fa14b6f7073.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "9ea0e9468316fd465cf1cd70353cefbd1320b3ab5da20b64bde69162f4789750", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d9669d1161e583f3931ec175566e47da26910af761f565616e1a6b0bb251cecb", "generation_seconds_batch": 21.678142830030993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.340320587158203, "genuineness": 4.162164688110352, "identity": 0.7979172468185425, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9ea0e9468316fd465cf1cd70353cefbd1320b3ab5da20b64bde69162f4789750.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "9ee59925be7a17d676299fd614744e56af11696c4d527d37669d52cdd992a43e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8bfc66990981d44b7c9cca71eb3ff8b8a27b40cd1b9069ece0fcdf3ad001305c", "generation_seconds_batch": 22.740851672017016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.41180419921875, "genuineness": 4.400654315948486, "identity": 0.5204821825027466, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9ee59925be7a17d676299fd614744e56af11696c4d527d37669d52cdd992a43e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "9f73d44434a2303bfd9a45e12fcf53aa523f76b3924d62be5659c64bd39d1ba7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5ec48fd9e8c4603d29a7687cd396e362447cabea6bb80832345e97ef92cbac36", "generation_seconds_batch": 23.771759801078588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.425304412841797, "target_z": 0.9062498353198589, "ce": 5.816103935241699, "genuineness": 4.21237325668335, "identity": 0.7855510115623474, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/9f73d44434a2303bfd9a45e12fcf53aa523f76b3924d62be5659c64bd39d1ba7.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "a14ac8565bfffa8c30e092d3a6af15970e863a4632f7a9319f34427960603492", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fc6324010cc8e68aef0224480e3a25ffaa90c0e4dcee10eecc0448d2748705c5", "generation_seconds_batch": 18.97512251092121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.259467601776123, "target_z": 1.8281250494040424, "ce": 6.141753673553467, "genuineness": 4.595211505889893, "identity": 0.8520490527153015, "hypothesis": "J'ai trouvé la 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_WARM_reg_high/a14ac8565bfffa8c30e092d3a6af15970e863a4632f7a9319f34427960603492.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "a15643c1e937f85e655ef9e523175b49ab7db4817ef95fdc6a7b948151de6b76", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5eb4909cc21fd29d51afd257a41b1a6536f6fae5633be2553d73790675316a24", "generation_seconds_batch": 16.43219315900933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 5.998785495758057, "genuineness": 4.268857479095459, "identity": 0.5495725870132446, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a15643c1e937f85e655ef9e523175b49ab7db4817ef95fdc6a7b948151de6b76.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "011f55204b3bd81e4a8b942c43da7e97822ebb3d35b84f86872e13cc05d1ecb1", "generation_seconds_batch": 17.274609473068267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.6903349161148071, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere brug.", "wer": 0.09523809523809523, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a1cf2cc2d8a63f871eb97d6d469f400403bebb30b0827588517b1f9dc79d1b62.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "a21a92abe4986a6f1e3f73b7d858b181ce64a1ff8634a95062b6b3a6e57e7dce", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "584f2ead07a5d7cd5e076b2efcbc629a4331ad0a23f8e453b6c596710bd28d0d", "generation_seconds_batch": 15.801240448025055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0120463371276855, "target_z": 1.554687508234007, "ce": 6.078212261199951, "genuineness": 4.3504462242126465, "identity": 0.7923189997673035, "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_WARM_reg_high/a21a92abe4986a6f1e3f73b7d858b181ce64a1ff8634a95062b6b3a6e57e7dce.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "29bfa61a3f42bdf08c1e47dcf19533b1eec7d8e59f985e404d0c12909f44cbdf", "generation_seconds_batch": 17.764571076026186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.324435234069824, "genuineness": 4.507347106933594, "identity": 0.7215843796730042, "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_WARM_reg_high/a28d64671d4fe74638f99af5124c48e12dbba834cd57271bc2ffd2abc8e4e108.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "a2e1a349b8b52a9f13cd3863ab434e219ee748b382489ea9e43bb8a7cefc03ce", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "df809d1777ed86370f1c32fd64427d27e470559656b4238e012a72fee2f20e66", "generation_seconds_batch": 17.49666249193251, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.118083953857422, "target_z": 1.6718749505959576, "ce": 6.324435234069824, "genuineness": 4.168440818786621, "identity": 0.5830198526382446, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a2e1a349b8b52a9f13cd3863ab434e219ee748b382489ea9e43bb8a7cefc03ce.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "a2e7b27266fa106e03b71960d9eec0aca5cd21417fc8ca37502ba66157da0979", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a95943770b72f57f1fd75f0f8f953089b6a075a898d6f06bb3bfdd3f793724d8", "generation_seconds_batch": 25.204812011914328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.471254348754883, "target_z": 0.9570313940951235, "ce": 6.133810997009277, "genuineness": 4.262581825256348, "identity": 0.8370614051818848, "hypothesis": "Ik 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_WARM_reg_high/a2e7b27266fa106e03b71960d9eec0aca5cd21417fc8ca37502ba66157da0979.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "a3013547d06e5c906464f14885cc5a28b15f49a939e3f6f806a59d1225356c8c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ad18340cec10d8c57f98d459c13a47d280a722178a8d4f92c8ade54f711a9122", "generation_seconds_batch": 16.458869230002165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.395919322967529, "genuineness": 4.080576419830322, "identity": 0.39747560024261475, "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_WARM_reg_high/a3013547d06e5c906464f14885cc5a28b15f49a939e3f6f806a59d1225356c8c.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "a30a12b85c96ae3c0a3ee3b7d27fb079ec077fc7c41614b37e2540d98806f6de", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3b47d8200721020d7790ede0b718f5c21e1aa9def368b8305891d65376bb3806", "generation_seconds_batch": 22.219682855997235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 5.895530700683594, "genuineness": 4.193545341491699, "identity": 0.5716159343719482, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a30a12b85c96ae3c0a3ee3b7d27fb079ec077fc7c41614b37e2540d98806f6de.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "262d7eeb306c57c8c78458be38966a2ddfbd648d57b92f23f0b20e532a9a9f2a", "generation_seconds_batch": 23.46334828599356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 5.919358730316162, "genuineness": 4.6956281661987305, "identity": 0.741291880607605, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a3c7fe6c459adc59ef5444f0e3f0e2be2eda823b5dc5f920cac43c6347e2d203.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "a40e3b5a070ffb832bf586a0c0cbccc8e5d9bf4e7731754c229d216ce75b6ab9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fd13504df074d24f7ab976241407ec78907b8c4b9dca41942618816cfb15bbab", "generation_seconds_batch": 22.704661333933473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4641849994659424, "target_z": 0.9492187047129612, "ce": 6.260893821716309, "genuineness": 4.400654315948486, "identity": 0.4860154390335083, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a40e3b5a070ffb832bf586a0c0cbccc8e5d9bf4e7731754c229d216ce75b6ab9.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "a43496582649b8ea75f948e03301ad58a44f1efb93e6dcaf7b89170d4f1bae60", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8d503015f7275cd8fb48e8e8721df9676e3d91b1b184401ee7c177ce9d51ed71", "generation_seconds_batch": 17.254701825906523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.24500846862793, "genuineness": 4.369274139404297, "identity": 0.6365736722946167, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a43496582649b8ea75f948e03301ad58a44f1efb93e6dcaf7b89170d4f1bae60.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "a4420936ae1f752869b52e54858e9c3c57e48f7aad2c18ada8334585e142f8fc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d46b42563719dc48f29d8a5cd3a8d357708ec6643965f359906d37b94077c751", "generation_seconds_batch": 23.86341965198517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9554929733276367, "target_z": 1.4921875741060635, "ce": 5.871703147888184, "genuineness": 4.3504462242126465, "identity": 0.6580127477645874, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a4420936ae1f752869b52e54858e9c3c57e48f7aad2c18ada8334585e142f8fc.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "57632a81b13c6ba7c7f755501cc5ed13a28e1b29e3023266533c5d6c2533495e", "generation_seconds_batch": 19.426016883109696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 5.982900142669678, "genuineness": 4.557555198669434, "identity": 0.7513632774353027, "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_WARM_reg_high/a4bb8464b102606fc33df8473999174b51e39c97b417935886cf6bcc999f4fd1.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "a54bc3a0a71e5c75f3bd6ac54beada9684aadb28db04032724e2823b05c6b5c4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "6a1c15d2e8f4aa24391470cdd5f6e2e603a3f06a3de6ade54bfa88ca2f988f27", "generation_seconds_batch": 17.02996214001905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.160499095916748, "target_z": 1.7187500329360281, "ce": 5.732706069946289, "genuineness": 4.017816066741943, "identity": 0.6715278029441833, "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_WARM_reg_high/a54bc3a0a71e5c75f3bd6ac54beada9684aadb28db04032724e2823b05c6b5c4.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "a5570621b83c3cb1f56c5d9253f8858eaeb73ed2df45524d96429bdd65383d16", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1e06ab92e0955a1834f9bfe27ceb39bf25b78b8cb405f59d6eefa29e176efc26", "generation_seconds_batch": 24.955271175014786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.6354551315307617, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in klikken.", "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_WARM_reg_high/a5570621b83c3cb1f56c5d9253f8858eaeb73ed2df45524d96429bdd65383d16.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "a57de02e6f8a58e151404665ed765589ce9690597fb34d90b0640d45a9acedad", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0332f434e4e58e446652c984cfaeaeaf62d6115351d195f474d98c21bcbdb80d", "generation_seconds_batch": 17.569233587011695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.202914237976074, "target_z": 1.7656251152760987, "ce": 5.855817794799805, "genuineness": 4.005263805389404, "identity": 0.6655576825141907, "hypothesis": "De deur 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_WARM_reg_high/a57de02e6f8a58e151404665ed765589ce9690597fb34d90b0640d45a9acedad.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "a591e2d1b913151b8df2069f9928175fe342cd227a4ce36cfce58a20e6102b9f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5c158b559a7d41c6b820854886e4ca3977d286b7a63796d3e3e1b0aba7daafdd", "generation_seconds_batch": 16.106708309962414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.7533392906188965, "genuineness": 4.507347106933594, "identity": 0.7173852920532227, "hypothesis": "Ik 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_WARM_reg_high/a591e2d1b913151b8df2069f9928175fe342cd227a4ce36cfce58a20e6102b9f.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "a6394084072140fedfa41cb4d64f2d6f9a415ed26b36b55b33b9130cce897a99", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "355dad7b875610293108c6c6e95b7efb1665a7c63b55f80002cae0b664fc45dc", "generation_seconds_batch": 16.668196739978157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446512222290039, "target_z": 0.92968763997812, "ce": 6.229123115539551, "genuineness": 4.444586753845215, "identity": 0.6790046691894531, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a6394084072140fedfa41cb4d64f2d6f9a415ed26b36b55b33b9130cce897a99.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "a68629c794d7b95fad2d5e1a40806031e9b1a81062bff778ec9f4c7141f19e42", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "326b0625f0c65a0192b97e1814467fc2bf82b2b207fd1273b7b8a1bdad7f2ffe", "generation_seconds_batch": 25.700152426958084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.324435234069824, "genuineness": 4.262581825256348, "identity": 0.4068586826324463, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a68629c794d7b95fad2d5e1a40806031e9b1a81062bff778ec9f4c7141f19e42.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "a6b5a5f56a67ba3095ab8f746d54f24ec3fe62f4b18ba7e6f5be9eb544a339cc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6b97b296233b079c75770c73fec01ce40dad123a93f738694995c79ad8a5045b", "generation_seconds_batch": 22.72654182394035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 5.974957466125488, "genuineness": 4.319066047668457, "identity": 0.6707742214202881, "hypothesis": "De deur 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_WARM_reg_high/a6b5a5f56a67ba3095ab8f746d54f24ec3fe62f4b18ba7e6f5be9eb544a339cc.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "a6d8ac8713f30ce127c908da3bf8cef3f3874deb55587d6d935e843ee42522c7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6b6606961ae20d785580e2bd9c45aa9125d539b4805f537dfc1d33123366e94f", "generation_seconds_batch": 24.422171998070553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.0305562019348145, "genuineness": 4.570107460021973, "identity": 0.6896249055862427, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a6d8ac8713f30ce127c908da3bf8cef3f3874deb55587d6d935e843ee42522c7.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "a7b238b01841890e28871984bd0efec24a378371bccc2dd0faf281a6cf2d70a5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4aa57cd042fc2b159e2ed95a2f1796b73dd3ce47403521ecd628b13e59d2cbd7", "generation_seconds_batch": 23.621055867988616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.38797664642334, "genuineness": 4.519899368286133, "identity": 0.6705852746963501, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a7b238b01841890e28871984bd0efec24a378371bccc2dd0faf281a6cf2d70a5.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "a7c0bed2cfdbae07e946942cddc85c83a95d637564b9b6662d34e884b5676086", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "51ddbf10602af02d669459201d7d979d9572896ea0317521da6c60a1520033ab", "generation_seconds_batch": 21.542026336072013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7716941833496094, "target_z": 1.2890623929579084, "ce": 5.95112943649292, "genuineness": 4.074300289154053, "identity": 0.364221453666687, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a7c0bed2cfdbae07e946942cddc85c83a95d637564b9b6662d34e884b5676086.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "a7e7ff0e25555607e3019bc764b5cc410c4d1d9068aa6e7a4dcfa08c51c7685a", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8be58be8cc9948944f57b534073c1117d9a03f7ee01d9136787bb481e00bfdaf", "generation_seconds_batch": 16.228829914005473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.125868320465088, "genuineness": 4.444586753845215, "identity": 0.6388998031616211, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a7e7ff0e25555607e3019bc764b5cc410c4d1d9068aa6e7a4dcfa08c51c7685a.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "a88052e5acc2914d78db092327ef073b47e9f5705790e6865b399f7c85a730c0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d1987dce965ea907e530b107beb1998f601c0e16103933b6e1d65aca58b0d71a", "generation_seconds_batch": 24.09594844689127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 5.967014789581299, "genuineness": 3.923675537109375, "identity": 0.541068434715271, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a88052e5acc2914d78db092327ef073b47e9f5705790e6865b399f7c85a730c0.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "a88c13a88260dcd86d30f6671a86b40dadb4dc7fbb42be62ad083dad5a85a5df", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b84e268320c4b0293536f6c2edd69a61fe5868cab26071857e67029db8e46e84", "generation_seconds_batch": 26.145704962080345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 5.244231700897217, "genuineness": 4.419482707977295, "identity": 0.7427455186843872, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble the cell contient.", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a88c13a88260dcd86d30f6671a86b40dadb4dc7fbb42be62ad083dad5a85a5df.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "a894bf4b13a79eee4c685b78b8ead6fd40d0d2261c3bbd90afa26a73d20ae036", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cf7b9c11fe3ae2a2b2279cbea2efdc7675019dc839c13e14ee7192a4b6eeeeb5", "generation_seconds_batch": 18.456865523941815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.6186484098434448, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a894bf4b13a79eee4c685b78b8ead6fd40d0d2261c3bbd90afa26a73d20ae036.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(very warm vocal delivery. 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": "500b868863235eba44859e7d598b76697825b647bc83493023b01c73f33be5b4", "generation_seconds_batch": 17.67209276801441, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.414700746536255, "target_z": 0.8945311964789542, "ce": 6.165581703186035, "genuineness": 4.250029563903809, "identity": 0.7206016778945923, "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_WARM_reg_high/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "a935a1ddbdb78c3471b07145898d37b2b0ebd6136c9e38b5b07f97113d959b25", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f5dcb0ffc8f42b0f4dc5ac4f3cefadb766512c494806db4f8034381f8a0f4989", "generation_seconds_batch": 23.151764514972456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.38288950920105, "target_z": 0.8593750164680141, "ce": 6.213237762451172, "genuineness": 3.8546390533447266, "identity": 0.2090068906545639, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/a935a1ddbdb78c3471b07145898d37b2b0ebd6136c9e38b5b07f97113d959b25.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very warm vocal delivery. 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": "0522d109d306efd39ddaf7fc1b8fb2ffde1c8e130211ca7eb1a924d2ed8db02d", "generation_seconds_batch": 18.211912594968453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.202914237976074, "target_z": 1.7656251152760987, "ce": 6.1894097328186035, "genuineness": 4.582659721374512, "identity": 0.8470662832260132, "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_WARM_reg_high/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "a98f1cfd77801bd4f8874a81653b7a12620cbc3c1d8c7d9453ccb7b886a0bd2f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3d56c047dda8a88eb2cb2a31503108916c82eff2daecff70961c5577213e4b89", "generation_seconds_batch": 16.19360716897063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729279041290283, "target_z": 1.2421873106178378, "ce": 5.752562999725342, "genuineness": 4.024092197418213, "identity": 0.7104988098144531, "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_WARM_reg_high/a98f1cfd77801bd4f8874a81653b7a12620cbc3c1d8c7d9453ccb7b886a0bd2f.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "aa26f880aea7441c75b073db1296419a3a8fc2b9dd2d36d6a1ccbbd02ce2d8fa", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "36a58333224cae52712a8d145c101ebc86b30a397e4efe7e4c5e91ce69f866c5", "generation_seconds_batch": 24.08157822699286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4606504440307617, "target_z": 0.945312491765993, "ce": 6.340320587158203, "genuineness": 4.281409740447998, "identity": 0.5549716353416443, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/aa26f880aea7441c75b073db1296419a3a8fc2b9dd2d36d6a1ccbbd02ce2d8fa.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "aa5ecc5002332f63ca9f425d646398a98276f16bfa4a4036f3c700633e5b5f93", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1476411c1bd56cf53801c9eb635531a67ebb4ffb93014bc3c46e446f91e03951", "generation_seconds_batch": 16.913906328030862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.213237762451172, "genuineness": 4.3504462242126465, "identity": 0.5581716299057007, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/aa5ecc5002332f63ca9f425d646398a98276f16bfa4a4036f3c700633e5b5f93.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "aaa09c62a5e3507c890336b353c6c51b99d0f0f20017970063c0b79d584fe0a8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b2efd4a95cc2b7d90ce1f30a4ab6d12c4ade782f7058210f1cfe448a4522ae1b", "generation_seconds_batch": 22.168349427054636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.5387191772460938, "hypothesis": "Die Tür 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_WARM_reg_high/aaa09c62a5e3507c890336b353c6c51b99d0f0f20017970063c0b79d584fe0a8.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "aaab67441efd7c559bff5f244c905838c501a9e80450bd66467add7774334bfb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "da76e901daa8edd792b06f3c10ff6b71ef1aead37dfeee834efa7789ecdffc06", "generation_seconds_batch": 25.33071229699999, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.188776016235352, "target_z": 1.7500002634882257, "ce": 5.998785495758057, "genuineness": 4.319066047668457, "identity": 0.7290922403335571, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/aaab67441efd7c559bff5f244c905838c501a9e80450bd66467add7774334bfb.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "ab267759f04bb83ceffd857904113f6c0a77c6137df2a2d6e6f4c4f0906d4ffd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "0f93a8e7c3db31b667353f761a6a4818c1c2f87f408c500a178657ad9a663d0c", "generation_seconds_batch": 22.48642150394153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9696311950683594, "target_z": 1.5078124258939365, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.537034809589386, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ab267759f04bb83ceffd857904113f6c0a77c6137df2a2d6e6f4c4f0906d4ffd.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "ab6019a24e4e8ee886d3affa03a1d4824f1f3384d03b7b4755cf460e20322ce6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "75d749d14d94fc4274c9c3cb475a0abcf27f7b97b62fe054f33778df20fad51d", "generation_seconds_batch": 23.449130148044787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 5.903473377227783, "genuineness": 3.9299514293670654, "identity": 0.7161548137664795, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ab6019a24e4e8ee886d3affa03a1d4824f1f3384d03b7b4755cf460e20322ce6.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very warm vocal delivery. 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": "a7eefcd76388cf59521599a84055a657f348685776c0ef36d862cd1673e04521", "generation_seconds_batch": 19.319175221957266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2768518924713135, "target_z": 0.7421875741060635, "ce": 6.499174118041992, "genuineness": 4.168440818786621, "identity": 0.3180682361125946, "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_WARM_reg_high/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "abf239b73b5259133eeb914933543fd8a07ea0d47d406438ccd1a17cfc380a05", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "95b947d34dfaa3fe6b85d98ad172393757352274301eaf30f01d121a9840562a", "generation_seconds_batch": 23.409014391014352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.8358935117721558, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/abf239b73b5259133eeb914933543fd8a07ea0d47d406438ccd1a17cfc380a05.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "ac91c3f59e89b9e6d3a2d5fb0a3fc1d97f7abf70bb6e397a651c5dbd8823c995", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2c639de8fec8114caa47714afbe6b0b9860554c7b2e231cf37e2c945565f9561", "generation_seconds_batch": 16.675720482016914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7151408195495605, "target_z": 1.2265624588299646, "ce": 5.935244083404541, "genuineness": 4.2751336097717285, "identity": 0.5387356281280518, "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_WARM_reg_high/ac91c3f59e89b9e6d3a2d5fb0a3fc1d97f7abf70bb6e397a651c5dbd8823c995.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6b18eac2ac5cefe400c89be7648a93488221774d69319ca379ad7f592607ff5d", "generation_seconds_batch": 21.073604903067462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.117925643920898, "genuineness": 4.482243061065674, "identity": 0.6489872932434082, "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_WARM_reg_high/ace1b3828dcfd65b325a43667842af365bcf502d6173fdb9904af1085b1569b0.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "ad6011d39635fef8a8105d0ceea5cfd1b66457b90ce9c54f80cf1c5e1a41ecac", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "3b175e6b6546d27f3b6f97c8d84d1f8836f045fe731b99f2b32749af933ddd27", "generation_seconds_batch": 26.659380369936116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.173524379730225, "genuineness": 3.9550557136535645, "identity": 0.7333716750144958, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ad6011d39635fef8a8105d0ceea5cfd1b66457b90ce9c54f80cf1c5e1a41ecac.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "55836461429154e4b0a6c50c2b99444c079f013042fe9fdaf9ec5df4ad4cfe30", "generation_seconds_batch": 21.225058619049378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729279041290283, "target_z": 1.2421873106178378, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.6621549129486084, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very warm vocal delivery. 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": "a120cadcfcc8a82a8c8796b7722034358b926ea37910d2f79d09f6099cf9a653", "generation_seconds_batch": 16.781732622999698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.117925643920898, "genuineness": 4.300237655639648, "identity": 0.7293135523796082, "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_WARM_reg_high/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "adfaa1d8b1cab08b318980be04ff52568537e5f8aa5c3a5c9ff397abbde99f7b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b210abaf256aec4540a45e79add182a1b8915e10b145513b77585cb5f6eb6dd", "generation_seconds_batch": 21.559644137043506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.260893821716309, "genuineness": 4.293961524963379, "identity": 0.6283529996871948, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/adfaa1d8b1cab08b318980be04ff52568537e5f8aa5c3a5c9ff397abbde99f7b.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(very warm vocal delivery. 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": "cc9290e33f55ade952ea6866099c863b2f6192be3d6687be86d0ca1a1be0b59c", "generation_seconds_batch": 17.802157877013087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.332377910614014, "genuineness": 3.8671910762786865, "identity": 0.28957629203796387, "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_WARM_reg_high/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "ae7fdb66371b264ef7f71c9bf19f29eb560246193b48b1fdb4e92bddc5b0a726", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6c8598ccecd6fa3d8284acd5af2f3be98056bda83f18a9938f571a6d9f0a7782", "generation_seconds_batch": 17.517801219015382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.133810997009277, "genuineness": 4.15588903427124, "identity": 0.42283886671066284, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ae7fdb66371b264ef7f71c9bf19f29eb560246193b48b1fdb4e92bddc5b0a726.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f6ccc78bd2187133d67858d231fa1b6c15acfe25ef1055c2456c9885e6a49cd3", "generation_seconds_batch": 24.210018084966578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9696311950683594, "target_z": 1.5078124258939365, "ce": 6.316492557525635, "genuineness": 4.3881025314331055, "identity": 0.7088377475738525, "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_WARM_reg_high/af449a5b4ee3e47d7e0052d4bac17b3fa22de9f3f18b93ca5a94a1ab7ebbe860.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "af6e01602eb8c3cfa34f2f11818994261eb85c21bd39ee630a9a32006319d2b1", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f5c76a842acf431de9ea3f7b8a6fdc712e1383ae2b05b473b44a4380277c8f17", "generation_seconds_batch": 21.207994668046013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.149606704711914, "target_z": 0.6015625905740776, "ce": 6.117925643920898, "genuineness": 3.8546390533447266, "identity": 0.48841655254364014, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/af6e01602eb8c3cfa34f2f11818994261eb85c21bd39ee630a9a32006319d2b1.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "24ee15f34e51b8ad041d345b667d64229cfc4ebc8b3c3c0c3c5c4178e7aac125", "generation_seconds_batch": 21.156972769997083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5172038078308105, "target_z": 1.0078124258939365, "ce": 6.1894097328186035, "genuineness": 4.20609712600708, "identity": 0.7096846103668213, "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_WARM_reg_high/af702ba4f04bb2509cf546a649a5a526dbf08b38a65074087664ad2073824e99.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "af7229adf028f3f885d6d9e6c47185323731e5848ea0973055f3a23efbee422e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "84d39f0311a51100feb359da37d5adc5a4ecb0fe2deceb3d32ba92f790d50a0d", "generation_seconds_batch": 22.31237850699108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3758203983306885, "target_z": 0.8515625905740776, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.6975461840629578, "hypothesis": "De deur 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_WARM_reg_high/af7229adf028f3f885d6d9e6c47185323731e5848ea0973055f3a23efbee422e.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "af841c5927af372b4d1f8c1e2116014e19ee9e3c6167d0d0cd2ff3a2d7f328ac", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b0cee624a5ee0fc26a20b6540e786f19c12e08d300285e2df54a3f626aa9cb75", "generation_seconds_batch": 16.38609918905422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6373798847198486, "target_z": 1.140624983531986, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.7190749645233154, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/af841c5927af372b4d1f8c1e2116014e19ee9e3c6167d0d0cd2ff3a2d7f328ac.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "b020d9e4920016bb6bd922a2ef0d1f556dbc31331f6042c26da1c72c0e92e421", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "954b3b1b2da3be41ea2a8869543ceefc807bd2602a994279b2cf75f4246c93df", "generation_seconds_batch": 18.579376109992154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.046441555023193, "genuineness": 3.9048471450805664, "identity": 0.5513864755630493, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b020d9e4920016bb6bd922a2ef0d1f556dbc31331f6042c26da1c72c0e92e421.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "b152b36e758c951876ec185eccc7275c8690a0d87ca55a62df72747b3020c886", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "51397a1897b48f138b3951bf6fc5f1d93ffc742f489113617cf1b70920e6188f", "generation_seconds_batch": 17.409799593966454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.828247547149658, "target_z": 1.3515623270858519, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.7407614588737488, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b152b36e758c951876ec185eccc7275c8690a0d87ca55a62df72747b3020c886.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "b1696c8ac061268f76cbc9961a13f5aa3e60a581c4703240c98cba15eaeabd44", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2e014296be1fcaeb79ec2f09f5bc326843a07978918146107b2b654666a88479", "generation_seconds_batch": 16.5599813190056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.197352409362793, "genuineness": 3.9801597595214844, "identity": 0.37726831436157227, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b1696c8ac061268f76cbc9961a13f5aa3e60a581c4703240c98cba15eaeabd44.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "b1c23b86a89fda58f1bc1161f551a1526994f4d273f20fe1989f8ee07a1943c3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f5d6ae28e5ed4010130f758faabc8ff87b1ad03749ca0d20dcaabe1f90cd5af3", "generation_seconds_batch": 21.53973842598498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 5.974957466125488, "genuineness": 3.961331605911255, "identity": 0.4766803979873657, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b1c23b86a89fda58f1bc1161f551a1526994f4d273f20fe1989f8ee07a1943c3.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "b1c709c80fe458f761c64063e7f7809309e4924f308836dcc7cbf3d4521439a9", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "33e42098b6ca62018a9ef0523f4c146a61580bc2175376d1867398a509d05601", "generation_seconds_batch": 16.19360716897063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.157639026641846, "genuineness": 3.71028995513916, "identity": 0.5696364641189575, "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_WARM_reg_high/b1c709c80fe458f761c64063e7f7809309e4924f308836dcc7cbf3d4521439a9.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "02216478c1464ebc139fc3993d2a6dbc1dd690465f1a48bab65fb7026896b5e5", "generation_seconds_batch": 17.815294245956466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.308549880981445, "genuineness": 4.231201171875, "identity": 0.6680302619934082, "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_WARM_reg_high/b1fe6ca3a04ef97497a3dd9dd4519f6940d468eb7ce388ea45a0df310b04a595.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a7e20e2d0ed9a9b582bfaaf7147ed0de9738a4de0faa3201aed88740965152c9", "generation_seconds_batch": 16.757155289989896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 5.959072113037109, "genuineness": 3.8797430992126465, "identity": 0.7776741981506348, "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_WARM_reg_high/b245f49daec2e5a04c40d20aba2ebedbf771b7154d20363c262d2d1ad08b863d.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "b24b26127dcad2b59621d644ac2b8fdd550b89f8aa7d94586add25d8ad8946f5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fee25cf9c9f996642ca808236b9469e4004c5f3c571ef50a07d9652959a0bcd3", "generation_seconds_batch": 15.555132296984084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.050638198852539, "target_z": 0.4921875741060635, "ce": 6.006728172302246, "genuineness": 4.118232727050781, "identity": 0.749735414981842, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b24b26127dcad2b59621d644ac2b8fdd550b89f8aa7d94586add25d8ad8946f5.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "b25f1c98a9de44fc331c83220d40a8cefc88e3d52571a30754871b8bb99a6835", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "e0d663f849de37ea7d15dbc016d7d5b6c95ec7eb9beb3bac5c9b3317bb0be5d9", "generation_seconds_batch": 22.412906074081548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.259467601776123, "target_z": 1.8281250494040424, "ce": 6.165581703186035, "genuineness": 4.6203155517578125, "identity": 0.7701773643493652, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b25f1c98a9de44fc331c83220d40a8cefc88e3d52571a30754871b8bb99a6835.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "b29b5716fd2007df881788ab9551b5f389e5c32fbb9c0ce7cd5a90552c2fdefc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c7eee20fe9bec08f0cdbf42b0818e15d378d25aba21859f6aa458d507f05aa09", "generation_seconds_batch": 23.58497692400124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.096876621246338, "target_z": 1.6484376729141481, "ce": 6.300607204437256, "genuineness": 4.218649387359619, "identity": 0.282940536737442, "hypothesis": "I found the little box. It was beside the old window. We can look inside to", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b29b5716fd2007df881788ab9551b5f389e5c32fbb9c0ce7cd5a90552c2fdefc.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "b390fa8819271615f0e17ffef1955287b4c0deb83591ec5f4b085d87120328ef", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "053e3449a2ffbed3f5fc31fd9716d20a3c2085f2bcc7e5440393fcf30021b9ff", "generation_seconds_batch": 25.097747804014944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.316492557525635, "genuineness": 3.936227321624756, "identity": 0.5237192511558533, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b390fa8819271615f0e17ffef1955287b4c0deb83591ec5f4b085d87120328ef.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "b3b1cb5bdf6d2ccc1d0ac5dca3dd26bf99ba65fbb739332549d0ed10cd9949c6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b35bdac0803ec02e37c99730261db42f7a18e18e43931e14905b2f6fbda1261d", "generation_seconds_batch": 17.360030864016153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 6.078212261199951, "genuineness": 4.18726921081543, "identity": 0.5819063782691956, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que antes.", "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_WARM_reg_high/b3b1cb5bdf6d2ccc1d0ac5dca3dd26bf99ba65fbb739332549d0ed10cd9949c6.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "b402fe6d42c1e346b8f6a76e54814df6966bd0763481200c4092d19dc1ed9fc2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cb8fc424deede6d1b0c29d222b158d05871fe9245c89be043e863038b8241a15", "generation_seconds_batch": 23.91956996207591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.040323257446289, "target_z": 1.5859377387862046, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.7921228408813477, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b402fe6d42c1e346b8f6a76e54814df6966bd0763481200c4092d19dc1ed9fc2.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(very warm vocal delivery. 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": "cf43860d854762890d5bc5e6c68d49c94861054b186f40c97dbdbb6f7156e447", "generation_seconds_batch": 18.52323825797066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.308549880981445, "genuineness": 4.15588903427124, "identity": 0.6116697788238525, "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_WARM_reg_high/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "b454a3122788f660102145157d34b548769489d2deb6270fb8f4d8f05e8b0f22", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5ffa6990fb61c053ef63b10a5b19d8c3742ee4627c2e9bfcff9c53d93b9ad568", "generation_seconds_batch": 16.986312270979397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.022613525390625, "genuineness": 4.519899368286133, "identity": 0.7895813584327698, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b454a3122788f660102145157d34b548769489d2deb6270fb8f4d8f05e8b0f22.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cf5595aad67632c477f900a863faf8dae46e3f34756b9f428055c94dd2ecc84a", "generation_seconds_batch": 24.13032099197153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.38003396987915, "genuineness": 4.174716949462891, "identity": 0.10730090737342834, "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_WARM_reg_high/b465393f79e18b9989a76b394ca786c7489f24626729fb5f2f0fe5962af057b6.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "b4c8c4291ecf7f1422d84968b113525e675761c1af9033a4d2aad2922aabe8f3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4adc7efa4f91722d8b7782ba385e3a019d9b03c217c865bd9a5ddcf11378d68c", "generation_seconds_batch": 24.758669962990098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.459460258483887, "genuineness": 4.444586753845215, "identity": 0.35276344418525696, "hypothesis": "I found the little box it was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b4c8c4291ecf7f1422d84968b113525e675761c1af9033a4d2aad2922aabe8f3.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "b519dac3e4e3cc747bc97a27b3da83d5465f932504694ea9d93feb5ca12c49fc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "610f16a9478bb7691d5cd8bab60771c81b0f0081363cd9cf42a8110987704d9c", "generation_seconds_batch": 23.718183303950354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7716941833496094, "target_z": 1.2890623929579084, "ce": 6.300607204437256, "genuineness": 4.20609712600708, "identity": 0.6344491839408875, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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_WARM_reg_high/b519dac3e4e3cc747bc97a27b3da83d5465f932504694ea9d93feb5ca12c49fc.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "b567245bc5da41f3ef35eaa4339db487a41d1017188830b7bf836f44c03f230a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9d7744edf9129a2f0429c6bc97ead97bb9ca97289fdd6506bf8f6f07b6c26745", "generation_seconds_batch": 25.92831645498518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.499530792236328, "target_z": 0.9882810976708695, "ce": 6.23706579208374, "genuineness": 4.218649387359619, "identity": 0.7766925692558289, "hypothesis": "Ik 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_WARM_reg_high/b567245bc5da41f3ef35eaa4339db487a41d1017188830b7bf836f44c03f230a.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "b5d0e2760e50c7d6da932f5e4d8d7a068f6763ed08083cab2a92c944368f47ed", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0a446cf3c6c523ddc3560f3a4f0987378e52af2ae8ac2b518cd07d13401ab6ed", "generation_seconds_batch": 16.255138618987985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.308549880981445, "genuineness": 3.9550557136535645, "identity": 0.3374940752983093, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b5d0e2760e50c7d6da932f5e4d8d7a068f6763ed08083cab2a92c944368f47ed.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "b67e2998bd2704720c6e84678e39a6746d99a35c61b38a64935c8de426495b8e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e21a284dbad2441616b80a51532a022dd814be209d71938556658108be0db29b", "generation_seconds_batch": 16.60803925804794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089807510375977, "target_z": 1.6406252470202116, "ce": 6.173524379730225, "genuineness": 4.15588903427124, "identity": 0.6854225397109985, "hypothesis": "Ik 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_WARM_reg_high/b67e2998bd2704720c6e84678e39a6746d99a35c61b38a64935c8de426495b8e.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "b68a0851c4f175d93dddf72d5f3a6cc6aed1422312b3e7ba976d555d1e0de8ed", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4a01fb9c3b74b5ee583a263ac5231bafee8f1d4f3be1f6c24c5c8f5299a429fb", "generation_seconds_batch": 16.913906328030862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686864137649536, "target_z": 1.195312491765993, "ce": 6.205295085906982, "genuineness": 4.124508857727051, "identity": 0.7693583369255066, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b68a0851c4f175d93dddf72d5f3a6cc6aed1422312b3e7ba976d555d1e0de8ed.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "b68c2d564081f1a0cfb3ff87a740c7044722785751f4a596719f9e4de3b51919", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f525e2af970888ec7c2281636e7b34a5df7558189c225ff7c1e7fdcec4927740", "generation_seconds_batch": 21.18986849905923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.337228775024414, "target_z": 1.914062788190247, "ce": 6.157639026641846, "genuineness": 4.457139015197754, "identity": 0.5777766704559326, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b68c2d564081f1a0cfb3ff87a740c7044722785751f4a596719f9e4de3b51919.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "b6f9495e9dcfa8af02d2f0a5e0d897334c63f68d3a05ce615e2a1079a3b70112", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "50e610841a9fe81e59be5d6fca045869c8bbe45669167d63dc97b1727a9a7931", "generation_seconds_batch": 22.31618132593576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 5.879645347595215, "genuineness": 4.65797233581543, "identity": 0.7132991552352905, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b6f9495e9dcfa8af02d2f0a5e0d897334c63f68d3a05ce615e2a1079a3b70112.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "b76bd2804b0a1f428d9e180a25f83d7afc960c7a6fb53f52bdf8646fa851a0d7", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8432b43840dd3df50acd7ef94559f15a22e371a4c78a8f3aa0045f629bd40f94", "generation_seconds_batch": 16.09815354901366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.326336145401001, "target_z": 0.7968750823400705, "ce": 6.141753673553467, "genuineness": 4.13078498840332, "identity": 0.7609520554542542, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b76bd2804b0a1f428d9e180a25f83d7afc960c7a6fb53f52bdf8646fa851a0d7.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "b79b74272a4e9675d05fd78d866b7c24e9da6389a83af29c2fc24c8460ad6acd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7a8d14e837ff9f7e597b014f9370a69246d44788dcce741641bc182419884e03", "generation_seconds_batch": 24.481939404970035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.205295085906982, "genuineness": 4.469690799713135, "identity": 0.7286130785942078, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b79b74272a4e9675d05fd78d866b7c24e9da6389a83af29c2fc24c8460ad6acd.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "b7cf9a33a3ca2773a1ba236b1faaa65a033d1bf72f9c82edf9ef1f1076d36afd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "b8838d810a355ca7e190eddc88bf96508f193326f4bf87baae45e960bad1551c", "generation_seconds_batch": 18.56928878207691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.205295085906982, "genuineness": 4.18099308013916, "identity": 0.2558072507381439, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b7cf9a33a3ca2773a1ba236b1faaa65a033d1bf72f9c82edf9ef1f1076d36afd.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "b7f34ee99436e1df728c50a965e2ba4bf03e51f37373e1c443465a9c83cbf9aa", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4acea29894be8d8240feb0058653782bf9cefc5a8556f6a882a067f4cd946a50", "generation_seconds_batch": 26.567971289041452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2311906814575195, "target_z": 1.7968748188518449, "ce": 6.0305562019348145, "genuineness": 4.356721878051758, "identity": 0.6792949438095093, "hypothesis": "Y he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b7f34ee99436e1df728c50a965e2ba4bf03e51f37373e1c443465a9c83cbf9aa.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "b7f527548bcb4fdf7ca2a2f5074e550a66c2bd9564e762d6a7404e43860ceaea", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5836a6ed894a56450f6169fcec159928241a15b400908ad4c2ce49c6b9c2049a", "generation_seconds_batch": 21.854197255917825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.157639026641846, "genuineness": 4.162164688110352, "identity": 0.5892802476882935, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b7f527548bcb4fdf7ca2a2f5074e550a66c2bd9564e762d6a7404e43860ceaea.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "b80aabd5020bad48964e61b608ad3fcc6049dae8c8c026af9967dd8c3dea6240", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "105678481fbc0dae63c6c5ca9e221de9a5e267ad79d1ff2377f707ccf6554425", "generation_seconds_batch": 21.437781027983874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.23706579208374, "genuineness": 4.2374773025512695, "identity": 0.11980746686458588, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b80aabd5020bad48964e61b608ad3fcc6049dae8c8c026af9967dd8c3dea6240.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "b82e170ba3b28460f3daa517f8e5db3263587860402c370d3f60ae208c71ef3d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d294f878b8fa74594a19ed8c35c8ec6086ee50f2cea09bd9cd9d2dcbce1b5f71", "generation_seconds_batch": 21.207994668046013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.268836498260498, "genuineness": 3.8797430992126465, "identity": 0.25983941555023193, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b82e170ba3b28460f3daa517f8e5db3263587860402c370d3f60ae208c71ef3d.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "b83e107eb20d9241eebf2f9a585568e255fabc8fec52c7c1542bbbbc8b24064e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "c73ca29d7c61eb386e609ae2fac68834af4305794f96c4ded6efaceaf6f2d61a", "generation_seconds_batch": 18.187444249051623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.252951145172119, "genuineness": 4.005263805389404, "identity": 0.5534687042236328, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b83e107eb20d9241eebf2f9a585568e255fabc8fec52c7c1542bbbbc8b24064e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "df469c45c2a46be11ca4e6a3751601612db0a7fd770d5ae469c6e9e81e00f452", "generation_seconds_batch": 21.47452757600695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 5.95112943649292, "genuineness": 4.068024158477783, "identity": 0.6961485147476196, "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_WARM_reg_high/b8c8380e21d9bed918870a379b131d40e5d4330b5822f423dea390bf6372b728.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "b8cbe331c1485050dc5031e404ed331a6c8a515a049cc9f267400b617269ad0b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ae8149da0640eb098b5eb0f29a418e76847e3780c0cb7defe367ae297298fb2a", "generation_seconds_batch": 24.605219756020233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.348263263702393, "genuineness": 4.400654315948486, "identity": 0.47490817308425903, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b8cbe331c1485050dc5031e404ed331a6c8a515a049cc9f267400b617269ad0b.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "b8cc8a20046bbe50bd1fbd4c4cbc00479cdc6c9ae809d3f9537bc1bd7c0defbe", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "839de7f3803820652ef731ee92815c4a8517bd7c8ff8efd3f8ab49fb47b1d75b", "generation_seconds_batch": 17.160005621961318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.217052459716797, "target_z": 1.7812499670639719, "ce": 6.395919322967529, "genuineness": 4.758388519287109, "identity": 0.5119259357452393, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b8cc8a20046bbe50bd1fbd4c4cbc00479cdc6c9ae809d3f9537bc1bd7c0defbe.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "b8d17b9120fec5ec9a589be4b066021226db8768abd50e3a52e3063a4b9f70e1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1851f712db14ab281870f592d1ad9a3f0d058e82315eeaad6193ff55025fb732", "generation_seconds_batch": 23.54294746601954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.268836498260498, "genuineness": 4.137060642242432, "identity": 0.5627519488334656, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfragen.", "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_WARM_reg_high/b8d17b9120fec5ec9a589be4b066021226db8768abd50e3a52e3063a4b9f70e1.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "b92d0568ab491e8416964320d390c770f2e04df7c25741b228dd8e93b4701502", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "29742ec92baf4f66a159d2188591045f1adf3308bfa3f0655e0f7f0303582910", "generation_seconds_batch": 22.145384954987094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.221180438995361, "genuineness": 4.15588903427124, "identity": 0.7688174247741699, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b92d0568ab491e8416964320d390c770f2e04df7c25741b228dd8e93b4701502.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "b94a53480e5bea81893ad32cf0f6f6d96334a620a98375f33ebb877ca80fc8b2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "224536dfd7b8ad0e753f2b9564de6c4ac3611f68efbbe017046d7d6ddf4f715c", "generation_seconds_batch": 23.97997178300284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.173524379730225, "genuineness": 4.381826400756836, "identity": 0.7288106679916382, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b94a53480e5bea81893ad32cf0f6f6d96334a620a98375f33ebb877ca80fc8b2.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "b9882325c072074c4aca9f72f94f92058fbf1624d8270c7bbe314ac6d01a8411", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9207623d2a86d8d0812458e73b1036e356d72fa344f267d8fbdc39834bc25dc4", "generation_seconds_batch": 24.897501318948343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 5.974957466125488, "genuineness": 3.9425034523010254, "identity": 0.4540719985961914, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b9882325c072074c4aca9f72f94f92058fbf1624d8270c7bbe314ac6d01a8411.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "b993ea9408c7dddba90bd69d6b98b8183217d3e9c1d59df854f42c16c0525f5c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "865c95698f70f30c1db10c77e03e9b687e956934189b35865573f3f7e14ed949", "generation_seconds_batch": 21.242770769051276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941354513168335, "target_z": 1.4765624588299646, "ce": 6.372091293334961, "genuineness": 4.3504462242126465, "identity": 0.6278461813926697, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b993ea9408c7dddba90bd69d6b98b8183217d3e9c1d59df854f42c16c0525f5c.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "b9979f0ca77e5add2f0a5e25b457370b0ae15a8621ee46328824f45a14c0799a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "753bfe83413c5d19785ff1c7b539d623571d042774fc42a5f0181ae87f5ba6bf", "generation_seconds_batch": 23.794667519978248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.372285842895508, "target_z": 0.8476563776271093, "ce": 6.157639026641846, "genuineness": 4.231201171875, "identity": 0.6320382952690125, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b9979f0ca77e5add2f0a5e25b457370b0ae15a8621ee46328824f45a14c0799a.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "b9e613c40ffbc9bb7c4eac3f87e2ef42e32b002b8d04ad28a6f33579cb75fe0f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "d2e1c174789aea57cb93f450d9666f3150c2bfe38335192c3c74111fca7b9564", "generation_seconds_batch": 21.432926325011067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344297409057617, "target_z": 1.921874687107732, "ce": 6.292664527893066, "genuineness": 3.9550557136535645, "identity": 0.3398284912109375, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/b9e613c40ffbc9bb7c4eac3f87e2ef42e32b002b8d04ad28a6f33579cb75fe0f.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c1af7d00ab3b4b7e767c38e8aa9686d75747da80740dfa9f3b98de0d5b1c1fa6", "generation_seconds_batch": 24.33590683399234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.006728172302246, "genuineness": 4.419482707977295, "identity": 0.5626187324523926, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek en daarna naast het raam neem ze mee als je werkte.", "wer": 0.18181818181818182, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ba245740af28ee045664bc7bdee66f8c6b2d8a01e98c7af2835d77ca01ecd130.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "ba27d12e3ac44e03451427628163e46eacade23e3eb24c8bf6dd654b556509ce", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "58b675166cd5e655977a1be6dbb7e7dba800d8aa8858c471cc0e27e13e431f0b", "generation_seconds_batch": 17.96937737299595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.312197685241699, "target_z": 0.7812499670639718, "ce": 6.1894097328186035, "genuineness": 4.331617832183838, "identity": 0.36183178424835205, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ba27d12e3ac44e03451427628163e46eacade23e3eb24c8bf6dd654b556509ce.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very warm vocal delivery. 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": "a09bc36e217c3ae70da059464d0065595b7738cbffbd5a74df694a08b834a3ce", "generation_seconds_batch": 18.633507664082572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.997908115386963, "target_z": 1.539062656446134, "ce": 6.284721851348877, "genuineness": 4.545003414154053, "identity": 0.6544678211212158, "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_WARM_reg_high/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "bb1f5fdb78b0347d5ab3600a32108a106424cac2bacc25bf4cabdef65c59ef35", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "c812d8fc04f54f19e6a9e8c90b9d0eafe1cac88afc6240b3208066ee511d0f2c", "generation_seconds_batch": 24.258861553040333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.308549880981445, "genuineness": 3.9550557136535645, "identity": 0.35282859206199646, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bb1f5fdb78b0347d5ab3600a32108a106424cac2bacc25bf4cabdef65c59ef35.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bb59716b224500e9d45e163fa0ffae571d50bf8dbbb9687764542cbfb1f52ef3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7e9b59155ae87365aa790c258dd8787adff69320060abdd5f6b132b808a445e8", "generation_seconds_batch": 17.7327950859908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.316492557525635, "genuineness": 4.331617832183838, "identity": 0.4828335642814636, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bb59716b224500e9d45e163fa0ffae571d50bf8dbbb9687764542cbfb1f52ef3.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bba4a145e52703a86b21bb6ccfaa53f1866193a58dc0faf9d8e034e19deec663", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "177d3ed4ca108fbb874799dd80fb8e77e7714543df8b9b7ae49ce5ca6af1d5bc", "generation_seconds_batch": 23.621055867988616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.701002597808838, "target_z": 1.2109376070420916, "ce": 6.284721851348877, "genuineness": 4.337893962860107, "identity": 0.4789677858352661, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bba4a145e52703a86b21bb6ccfaa53f1866193a58dc0faf9d8e034e19deec663.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bbf4ce3357d57210804a72368b8a921b0bbee4d30909deaf6c045e9e52c918c8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b914ffe1a9abebeb156702e71eb1eb768f6c7ca51bd8718132d202ae3ed58365", "generation_seconds_batch": 16.91247162304353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5596189498901367, "target_z": 1.054687508234007, "ce": 6.2767791748046875, "genuineness": 4.2751336097717285, "identity": 0.4563271105289459, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bbf4ce3357d57210804a72368b8a921b0bbee4d30909deaf6c045e9e52c918c8.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bbf647e7b0b29b39f318fc7ae70edf2619d56a6c8253ad40d92ee6af3b772eed", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8e21825c221d8e798ebe6006df2c18808283ffb0fa462d54f10f23764e9702a1", "generation_seconds_batch": 16.310697073000483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.7401689291000366, "hypothesis": "Die Tür 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_WARM_reg_high/bbf647e7b0b29b39f318fc7ae70edf2619d56a6c8253ad40d92ee6af3b772eed.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6d05eedd941f3b2d3258fb9fa41443e99f19583c368409ab83c427fd772a3523", "generation_seconds_batch": 20.426329857087694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.308549880981445, "genuineness": 4.413206577301025, "identity": 0.6138837337493896, "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_WARM_reg_high/bc8aff5144ab41203905b17340ca37fbf943d8517f7b728ea93b8a968fcad280.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "bca10b52102002e512c59838408610ce572ccb7a1731780e014b7e255be023ac", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b01272f5e069e6b49e4c4bc69fa80d71a63e538a078eead8b3406587da5d5281", "generation_seconds_batch": 16.38609918905422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.133810997009277, "genuineness": 4.13078498840332, "identity": 0.6314708590507507, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bca10b52102002e512c59838408610ce572ccb7a1731780e014b7e255be023ac.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "bca65137541d13296e4e1131ad4469a43218040763baae5e591a10a88ea202d3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8f60787f9658e8a54fa4ed6a4521c4266a3b2b017f7235a6e2bda28a0576537b", "generation_seconds_batch": 22.365867875050753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.7713236808776855, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bca65137541d13296e4e1131ad4469a43218040763baae5e591a10a88ea202d3.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "bcaecff7e56405548f05b7cddb6e2e4a65fc824ab6c1aa9b25f744f7fe385216", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f6aa2f27829f566a9a5deb80d4eb26c6b487c15c068436330ca5376cac3e9780", "generation_seconds_batch": 23.39201901596971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.141753673553467, "genuineness": 3.7165660858154297, "identity": 0.19417110085487366, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bcaecff7e56405548f05b7cddb6e2e4a65fc824ab6c1aa9b25f744f7fe385216.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "bcefd59b616828bb356298dd104d00f00f7c7d8048b8387b80e7121a2cdb8d89", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8b249303a97d14b687fe8b338914bf49f9908f268acbd52ba3d310d766870e6d", "generation_seconds_batch": 24.172303588013165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.284721851348877, "genuineness": 4.381826400756836, "identity": 0.7484632134437561, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in keijer.", "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_WARM_reg_high/bcefd59b616828bb356298dd104d00f00f7c7d8048b8387b80e7121a2cdb8d89.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "bcfd4cfe6b5d109b07dad18bae276b77578d47de39790bde18a523ccf3376d80", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d8f84005fb8572dea93286b24643ffb444feea2170d77ceab92b3fdf85442762", "generation_seconds_batch": 23.354315256001428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.332377910614014, "genuineness": 3.9801597595214844, "identity": 0.5729559659957886, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinkommen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bcfd4cfe6b5d109b07dad18bae276b77578d47de39790bde18a523ccf3376d80.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d6f7da9df5d2b63389161e57f7b894f525a5d4b781d7528c938aeb98a9c874be", "generation_seconds_batch": 18.1512148269685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.6822832822799683, "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_WARM_reg_high/bd38e0c515d9d04b47fd9c8de77e935c6927165e7286247cb51346eb1473486a.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bd4823a64e5ed4cb970ecd6a03c183a0d5974112b4bdcc030ad5ba1e41d58f82", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f7ad907997bc051a1f79ef5312017f47c4d10f20b92535cf01f64a022eb97b2c", "generation_seconds_batch": 16.174031116999686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.181467056274414, "genuineness": 4.18099308013916, "identity": 0.8272278308868408, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bd4823a64e5ed4cb970ecd6a03c183a0d5974112b4bdcc030ad5ba1e41d58f82.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "bd4e4a9b1eb176d6e773f5c876bf7ce278014b1d24de5859307cbc7bc32ed0f3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b81c212bec0d6b05ff8aa38066d32708da1710f1093d6be8198dbd516ecd3604", "generation_seconds_batch": 16.47887192806229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.109982967376709, "genuineness": 3.948779582977295, "identity": 0.4977513551712036, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bd4e4a9b1eb176d6e773f5c876bf7ce278014b1d24de5859307cbc7bc32ed0f3.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bd5d74915e368d5fa663a2f52e1ae4d0ecb37e024205ff29dc1bad1e50ba9a0b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6f464da0b443499e30b9ca016d27e9c7090dc5a96d5483ff3c44864edac44583", "generation_seconds_batch": 21.662563862046227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.268836498260498, "genuineness": 4.344170093536377, "identity": 0.5897274017333984, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bd5d74915e368d5fa663a2f52e1ae4d0ecb37e024205ff29dc1bad1e50ba9a0b.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "bd8dee1b89b63944439db975f807752a091fc5ae493d5677787281effdaa35dd", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "edc4aeb209f813b10f108f2a54a6cb101b6672abe42af8553e9bc21a937843ea", "generation_seconds_batch": 17.802628564997576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.260893821716309, "genuineness": 4.507347106933594, "identity": 0.5345762372016907, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bd8dee1b89b63944439db975f807752a091fc5ae493d5677787281effdaa35dd.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "be02b37abbc5d5b7f8ebe01b21c3c869040c4825a2f0ae93fb81ea0fe3257c2a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f9c728710064e48e58641cabd5548bfa4df49cfae6b1c9797f74400cc7719a83", "generation_seconds_batch": 20.859213036019355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.229123115539551, "genuineness": 4.268857479095459, "identity": 0.8073285818099976, "hypothesis": "Ik 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_WARM_reg_high/be02b37abbc5d5b7f8ebe01b21c3c869040c4825a2f0ae93fb81ea0fe3257c2a.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "be668e773e4b18e7d2de94ca86f82a8165a8094626f15d0788bd00122e8cec68", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "46fa351b5616d2de4fe9bd7dd06408df5e4d7d21a787d16ee302ed2fca96d057", "generation_seconds_batch": 19.30085990205407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.305128574371338, "target_z": 0.7734375411700353, "ce": 6.09409761428833, "genuineness": 4.287685871124268, "identity": 0.7314138412475586, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/be668e773e4b18e7d2de94ca86f82a8165a8094626f15d0788bd00122e8cec68.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "be7bef80a87ee51465a143deaa8aff2a12a0b1b1b7fa4e40ea87b010e280d174", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "45845302035edb002f4b763844eca9ea6c38a616abd44754f739b679bbb0db98", "generation_seconds_batch": 23.00109107105527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.696911096572876, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/be7bef80a87ee51465a143deaa8aff2a12a0b1b1b7fa4e40ea87b010e280d174.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "bf2f007449966391933b05428799ae947510737934cb34b73daed7d80ef17ddf", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cf934853377f67547429e92bc2906f9516b3145602e1bd4f22b650155d6ab76d", "generation_seconds_batch": 19.30085990205407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.389958620071411, "target_z": 0.8671874423619507, "ce": 6.24500846862793, "genuineness": 4.356721878051758, "identity": 0.7449910640716553, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bf2f007449966391933b05428799ae947510737934cb34b73daed7d80ef17ddf.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "bf4e00bbf9e3a77c1173f4fb4781cee7012cd271e52acaea35115786af89803b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "294326a80049b42fcbb823e7879860a8a7a1522020093fe51d78589219c95dd2", "generation_seconds_batch": 17.35024698509369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.47577327489852905, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bf4e00bbf9e3a77c1173f4fb4781cee7012cd271e52acaea35115786af89803b.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "bfb939bd863d2a7ff03ddef4d7bc48a77ced2c6f6e8c4f813e0ea22d98168d3a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "bdf4e92d702381d7bad9c6fee02d2f03d001077b3ccdf3252377bb1984e05755", "generation_seconds_batch": 21.40773731400259, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.117925643920898, "genuineness": 4.074300289154053, "identity": 0.30314400792121887, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/bfb939bd863d2a7ff03ddef4d7bc48a77ced2c6f6e8c4f813e0ea22d98168d3a.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "c03d8d03ee6c2442b3094b0d42ae252f3df71492b4563e4c54ef2cac02712371", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "35d96dd091f7588284c10981811db2a5c494204fd849fcf73b914165edb79af8", "generation_seconds_batch": 17.507038222043775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.324435234069824, "genuineness": 4.1998209953308105, "identity": 0.6371475458145142, "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_WARM_reg_high/c03d8d03ee6c2442b3094b0d42ae252f3df71492b4563e4c54ef2cac02712371.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "c15fbf037985698bda924e95583adf149448fdb0005a2549680a76d8120e2337", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d968bbe01521b9838fb33d74f94e5057ddb3290590cbf4629c89d8756fd82293", "generation_seconds_batch": 24.176765793003142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5172038078308105, "target_z": 1.0078124258939365, "ce": 6.308549880981445, "genuineness": 4.281409740447998, "identity": 0.5442860126495361, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c15fbf037985698bda924e95583adf149448fdb0005a2549680a76d8120e2337.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "aba493d7d25db42f3460aba752b45bf4b1df1be6045aa51d51f05bdff9895c75", "generation_seconds_batch": 24.13032099197153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.300607204437256, "genuineness": 3.8420867919921875, "identity": 0.4222516715526581, "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_WARM_reg_high/c16dfc72778a22c72d7ace57cca99dd6aeb62fc274ee6eebe765daf7a3973f98.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "c16e136518a33221b4177a395c753f7de643e174c1a096f26354664cde9947b4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f86aabadb49f05b1e07ff0b402e391e91b70a357306eed27843f05481104e79c", "generation_seconds_batch": 23.350336343050003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.173524379730225, "genuineness": 4.419482707977295, "identity": 0.6645116209983826, "hypothesis": "Die Tür 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_WARM_reg_high/c16e136518a33221b4177a395c753f7de643e174c1a096f26354664cde9947b4.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "c1b0ba84837a4abe8c28e540965d1efaa833e09a58c0ac4f8e2be9ddf3c0b971", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d29d8ccd6172a21fcbf91f3a1457a9fcd858a5e96c789b7bc849a94bc5db7fb6", "generation_seconds_batch": 22.86091861105524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4288392066955566, "target_z": 0.910156311755053, "ce": 6.109982967376709, "genuineness": 4.344170093536377, "identity": 0.7880969047546387, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c1b0ba84837a4abe8c28e540965d1efaa833e09a58c0ac4f8e2be9ddf3c0b971.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "c28e1caa0dabbacb1e96735c332cbc3845dc8844b1036de3fd1764fc23ade0e6", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d740f80794239698b6f80dfaf55b99a5b56d8d0e746c0e8766bf743f924e928e", "generation_seconds_batch": 16.61556583608035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 5.792276382446289, "genuineness": 4.005263805389404, "identity": 0.6441312432289124, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c28e1caa0dabbacb1e96735c332cbc3845dc8844b1036de3fd1764fc23ade0e6.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "c29dfeee24654633411aaf67c0a00616387cd739815bab0d6fb129349ba5b6a5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "15c3d95efce14c76f22c37f14819e72771babdf41df2db456421a9b64b2dbadd", "generation_seconds_batch": 18.075655404943973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0541727542877197, "target_z": 0.49609378705303175, "ce": 6.157639026641846, "genuineness": 4.21237325668335, "identity": 0.6907907724380493, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c29dfeee24654633411aaf67c0a00616387cd739815bab0d6fb129349ba5b6a5.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "c2a15bf7277eba246d24813f053b62f489f44472a205d713fdaee2134087861f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8f8b7cbbb410eb6c63371d11b4407bebe0375ef2935525a7fd9294d6fbb9ead5", "generation_seconds_batch": 14.402103102067485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.221180438995361, "genuineness": 4.337893962860107, "identity": 0.4703519344329834, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c2a15bf7277eba246d24813f053b62f489f44472a205d713fdaee2134087861f.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "c2a2c49fae74b7a440ae857b368005f94c75c97f90396ccec2865a8b569d4e47", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "01181bec9189360326dd3bef2d10a9102d513da543703982a25206e7f1061c64", "generation_seconds_batch": 23.351503244950436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316020965576172, "target_z": 1.890624983531986, "ce": 6.268836498260498, "genuineness": 4.419482707977295, "identity": 0.6935189366340637, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c2a2c49fae74b7a440ae857b368005f94c75c97f90396ccec2865a8b569d4e47.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "c2ab8896168075525b650912fbb6cf6dcc5de5ae9b6438354e91108667609fb5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "403ecbfd4f899d2284bde2bf644bb1cebdfab552d0f206df9db755f9c4683427", "generation_seconds_batch": 23.351503244950436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.157639026641846, "genuineness": 4.256305694580078, "identity": 0.8054971694946289, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c2ab8896168075525b650912fbb6cf6dcc5de5ae9b6438354e91108667609fb5.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "c2b2725a769d095909cc744370f836f814a6981447afd090ad494f8980f14d6b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "beaae87855053ccf1c7c91212e46b6b70b08e8d936bc57c692f5f2e2674b173b", "generation_seconds_batch": 23.967351846979, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.400851249694824, "target_z": 1.984375148212127, "ce": 5.839932441711426, "genuineness": 4.369274139404297, "identity": 0.8007168769836426, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c2b2725a769d095909cc744370f836f814a6981447afd090ad494f8980f14d6b.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "c340fcbfbb3c894c860065af100b82d42af7dfa720f5174f93345697afb2abeb", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e033205da4760022ab8f84e1fc48e4013bad535bfaa7135690d007d7e941e567", "generation_seconds_batch": 17.66231753793545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.316492557525635, "genuineness": 4.218649387359619, "identity": 0.2943122386932373, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c340fcbfbb3c894c860065af100b82d42af7dfa720f5174f93345697afb2abeb.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "c366762c019a3efe52ea3bf3fee0cdcb5b296254957f8aaa5191e9971f5d0a2d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f185da4f26ce72ebe0e1d1faa8e9ba887b4ea6fbe699358a8bcb20bcc47ef164", "generation_seconds_batch": 16.929879443952814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9767003059387207, "target_z": 1.515624851787873, "ce": 6.268836498260498, "genuineness": 4.7081804275512695, "identity": 0.7736701369285583, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c366762c019a3efe52ea3bf3fee0cdcb5b296254957f8aaa5191e9971f5d0a2d.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "c3880cbf5a2bdb8fe4e07c0bc7736597e3f14db8156c59d3334e280d383f15a5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "44f8506484ac2afeff4b0b1a73acfbe7f727392120231c3477e3b79e65aa0a00", "generation_seconds_batch": 23.278218221035786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.316020965576172, "target_z": 1.890624983531986, "ce": 5.700935363769531, "genuineness": 3.7542223930358887, "identity": 0.4898912012577057, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla, dime.", "wer": 0.2, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c3880cbf5a2bdb8fe4e07c0bc7736597e3f14db8156c59d3334e280d383f15a5.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(very warm vocal delivery. 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": "a7325ca0f40f73bec7eb587769daf6bde355a07e9046e9d8da89d3585655e5f0", "generation_seconds_batch": 16.301049164962023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.280386447906494, "target_z": 0.7460937870530318, "ce": 6.054384231567383, "genuineness": 4.268857479095459, "identity": 0.6029889583587646, "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_WARM_reg_high/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "c3a34b8e567b4c30dbf59b09796d37073fc981fb47fd7ea091d7b94671970b0f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5b288e2670f5bf9e718977deb825e475de6af8ff3066ff079ebf1bca4ef49b9f", "generation_seconds_batch": 22.782345279934816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743417739868164, "target_z": 1.2578126893821622, "ce": 6.324435234069824, "genuineness": 4.093128204345703, "identity": 0.6195085048675537, "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_WARM_reg_high/c3a34b8e567b4c30dbf59b09796d37073fc981fb47fd7ea091d7b94671970b0f.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "45c6434ffd5dbebfe443bf83a59824b56112dd0d84caeb10f21e584bb2fdde33", "generation_seconds_batch": 19.426016883109696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 5.760505676269531, "genuineness": 4.218649387359619, "identity": 0.7993357181549072, "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_WARM_reg_high/c48ab466bf0b9c671df9868fc2f1d941a5558f04d022a37759acb0749197d2ec.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "c48c2551b8b6ff6c6e836a1958780f57971e3eb298298bd16aed9f3ea7822bd7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "02785ec74eb5f1280c02acdec8ead03d486ccd5a0f5395db18a0b9dbeb178a7a", "generation_seconds_batch": 25.178820859990083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 5.855817794799805, "genuineness": 4.099404335021973, "identity": 0.6164470314979553, "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_WARM_reg_high/c48c2551b8b6ff6c6e836a1958780f57971e3eb298298bd16aed9f3ea7822bd7.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(very warm vocal delivery. 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": "efade001383a38fd887ccef3d76c9508f266c2626cf282ed46af1029f7a6aeb0", "generation_seconds_batch": 16.79091814206913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4182353019714355, "target_z": 0.8984374094259224, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.6863263249397278, "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_WARM_reg_high/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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "008494b89c65bc7fdabb2e52399fb4f0741363f6bf4090eb662b436d6ad593e5", "generation_seconds_batch": 17.144357090000995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686864137649536, "target_z": 1.195312491765993, "ce": 6.395919322967529, "genuineness": 4.049196243286133, "identity": 0.3779001832008362, "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_WARM_reg_high/c497ae2ca6d7a50b40b82fa995e0bc5e5f9cdd90aff36833962173350d12ca00.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "c51ebe5374fdf3d8d466b93e01bc1d205b24baa9518e273050d8a4dcb1fba1d3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7329e114f0df4704130dee874778dccf5c731b3c2da9965ddecb49eaf4a6a2e8", "generation_seconds_batch": 17.66451627900824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.103945732116699, "target_z": 1.6562500988080846, "ce": 6.300607204437256, "genuineness": 4.394378662109375, "identity": 0.6196380257606506, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c51ebe5374fdf3d8d466b93e01bc1d205b24baa9518e273050d8a4dcb1fba1d3.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d8d6c92949d0781657e0fb60a0407bb5be76bfe36539a536a3592b4e83540dd1", "generation_seconds_batch": 18.1512148269685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.026184558868408, "target_z": 1.57031236002188, "ce": 6.205295085906982, "genuineness": 4.419482707977295, "identity": 0.49501273036003113, "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_WARM_reg_high/c526c91a581543320c81960027a14982cb54e8649123e0b3ddfef9b04a3e88b7.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "c5363649e2e023f2df8cbf7dd723828166c6eb4bce0e38b9c06e30f9fedeab50", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d3fd59b271fec9c835ce4692620678284a9f52656ec8dbf7f2431d1bb2f348f1", "generation_seconds_batch": 17.311734304996207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.300607204437256, "genuineness": 4.545003414154053, "identity": 0.5043820142745972, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c5363649e2e023f2df8cbf7dd723828166c6eb4bce0e38b9c06e30f9fedeab50.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "c5ca099b37a73ced0372737a9732838f02f4f24d856cb43d460f801bae90fcee", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "c32b6f15137b444ff8bad289b5b7a6d5b69c732e1df63499417776593c2ceb5e", "generation_seconds_batch": 22.90902197198011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.38003396987915, "genuineness": 4.293961524963379, "identity": 0.4836166799068451, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c5ca099b37a73ced0372737a9732838f02f4f24d856cb43d460f801bae90fcee.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "c5e2eb063623b136ccc7a5019efdc98f692c70ac7e5d7624561e1eb030494bfe", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4b884109f813e0bd39f51b2d625068b8a7990fdd749a78eb1749f37bf80bb764", "generation_seconds_batch": 23.982555918977596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.35418272018432617, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c5e2eb063623b136ccc7a5019efdc98f692c70ac7e5d7624561e1eb030494bfe.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "c5ed4dbcb8ba0339e7cb95e1cfaa9b874fdaf1e716f08031ba6fb6f62a47e278", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "d3cde217d3ee7c0aa2a4a6d77bf07d7c226a3c61d52d2e26ebc7806f136fa17c", "generation_seconds_batch": 24.09494421002455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9130778312683105, "target_z": 1.445312491765993, "ce": 6.340320587158203, "genuineness": 4.344170093536377, "identity": 0.43737155199050903, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c5ed4dbcb8ba0339e7cb95e1cfaa9b874fdaf1e716f08031ba6fb6f62a47e278.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "c6bc517812a059d374fc85b2c031175f5bd77c837a87d576db52567cdc0fbce7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9bbfe949217f6eff967c8c5580d876fcbd0997c2e641fdd92a0ff85194875b1b", "generation_seconds_batch": 23.451818525907584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.173524379730225, "genuineness": 4.344170093536377, "identity": 0.4845346212387085, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c6bc517812a059d374fc85b2c031175f5bd77c837a87d576db52567cdc0fbce7.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "c703b66f36b166e3219fd287a148e28c8205c40233ad111a93e3e04feb92de13", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b3a16278c29df3b358f54272bf6fb26043d4adde9645ef3d739fa0cf37bacfef", "generation_seconds_batch": 18.428318747901358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 6.594485759735107, "genuineness": 4.758388519287109, "identity": 0.7535257935523987, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c703b66f36b166e3219fd287a148e28c8205c40233ad111a93e3e04feb92de13.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "c811f2d42b2bddadba20e33efd7ee28ed1f49c07b240a6746bd76d8c3350d7e5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "27803d9a9e2438ecfe0a151e25cbcac5a57433e69eb16f8f3c2ca169bf6d30a9", "generation_seconds_batch": 24.883227605954744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.006728172302246, "genuineness": 4.118232727050781, "identity": 0.5671852827072144, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c811f2d42b2bddadba20e33efd7ee28ed1f49c07b240a6746bd76d8c3350d7e5.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(very warm vocal delivery. 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": "d2067ab9522d9f8a44754b8fde8f7fdace7e51928e30331ecfb83dcd0b1733b2", "generation_seconds_batch": 16.781732622999698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.213237762451172, "genuineness": 4.469690799713135, "identity": 0.7031466960906982, "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_WARM_reg_high/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "c851e87bb4cecfc71537d5fd708a394ef86da547591599738555e1a6a3b20ba4", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6fac2c8202b9662707fa58a8fb3b51a141d6cfa38eaab499ab65cd51087d2feb", "generation_seconds_batch": 17.7327950859908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.23706579208374, "genuineness": 4.557555198669434, "identity": 0.6224887371063232, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c851e87bb4cecfc71537d5fd708a394ef86da547591599738555e1a6a3b20ba4.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "c88881a3bc94e209b99ff8973c588766a0fc3af873f64c4d1bb5d612d79d5661", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0ed7303abbda32e1e6c95b6f3216cdf00615bda2ac15254eff80b7fef7f151dd", "generation_seconds_batch": 22.961509054992348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.372091293334961, "genuineness": 3.8922953605651855, "identity": 0.1781928539276123, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c88881a3bc94e209b99ff8973c588766a0fc3af873f64c4d1bb5d612d79d5661.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very warm vocal delivery. 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": "85767f442ff23170dd0ecc5bae9a5641a6cb5d61162df2b09b8ad35f8f252f94", "generation_seconds_batch": 17.741529052029364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.41180419921875, "genuineness": 4.432034492492676, "identity": 0.4124448597431183, "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_WARM_reg_high/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "c8ae872020360d6de977e9aeac973807b187a1c3c1fa8b9dddc5a7dbbd62b086", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4b9a6ecba3c267b471ca08b5dea4aa4dd6a68c0e29b73969ae0a8272daa49ef3", "generation_seconds_batch": 23.959343201946467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743417739868164, "target_z": 1.2578126893821622, "ce": 6.070269584655762, "genuineness": 4.281409740447998, "identity": 0.48210716247558594, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c8ae872020360d6de977e9aeac973807b187a1c3c1fa8b9dddc5a7dbbd62b086.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "c8be8cd90a414aaecd876d27b98cd1400f4e2a29508ccd21d1037f98faf4e086", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f6ad295145b1843aba61901be9ff50c7b18bb5d65873ea22f762fb66a8fd1547", "generation_seconds_batch": 23.05154339398723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.23706579208374, "genuineness": 3.936227321624756, "identity": 0.5702651739120483, "hypothesis": "ik 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_WARM_reg_high/c8be8cd90a414aaecd876d27b98cd1400f4e2a29508ccd21d1037f98faf4e086.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "c92d5230df14879f6f10473c025bc8d5b58d48cfcd4a3a16c654169f6b2090b6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f4e863772909241496d9a8db90de5afbfc8117825be28e637da0880ab2c76ad3", "generation_seconds_batch": 24.09594844689127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.046441555023193, "genuineness": 4.105680465698242, "identity": 0.45610442757606506, "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_WARM_reg_high/c92d5230df14879f6f10473c025bc8d5b58d48cfcd4a3a16c654169f6b2090b6.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "c9b25bae30d8eb6502c68edd7215e20299e3adc9fed2f73458e7054fb95f1846", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f01337b3d86643fbc37fddfed1205cd12448e7dd1eca74df4a5be439514c91e2", "generation_seconds_batch": 24.542143414961174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174637317657471, "target_z": 1.7343748847239013, "ce": 6.523001670837402, "genuineness": 4.532451152801514, "identity": 0.6159734725952148, "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_WARM_reg_high/c9b25bae30d8eb6502c68edd7215e20299e3adc9fed2f73458e7054fb95f1846.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "c9c9edc1e5ec96dbcef9fcecb5657c9d53ceac02eaba94a5a1f1f1a78d9365e3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80b631cb17982e86bd7d51dc711ac6c516caa7ab794e9002bbd531334ee414bd", "generation_seconds_batch": 15.35720837302506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 6.332377910614014, "genuineness": 4.582659721374512, "identity": 0.40651559829711914, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/c9c9edc1e5ec96dbcef9fcecb5657c9d53ceac02eaba94a5a1f1f1a78d9365e3.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "ca3fff959c5ec8035eb3e68f09b378cc637e079712c68509a80bacaa700be60e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "19f5c763af0e8d4788258ee1956359196a4d42e81e6fadc4f3f7cda3b2567564", "generation_seconds_batch": 17.708350779954344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.332377910614014, "genuineness": 4.444586753845215, "identity": 0.6878260970115662, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ca3fff959c5ec8035eb3e68f09b378cc637e079712c68509a80bacaa700be60e.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "ca6a0af9e56aac0a80826b3feaaf79dbeec7e0043364cf6f869c971de5764170", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "cf120d402dc04416bedcd9e854e0956a5fdf9401b9b0ac13c0b141ac7915baab", "generation_seconds_batch": 23.439149915007874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.0146708488464355, "genuineness": 4.645420074462891, "identity": 0.6197389364242554, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "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_WARM_reg_high/ca6a0af9e56aac0a80826b3feaaf79dbeec7e0043364cf6f869c971de5764170.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "ca874c1dabf0677868e2b5e85a094aff611bde6a427fef50a32eacca6fdc5f80", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "45430c925e7aa1b861e5c10bf15cd203e4f4d1878c47162866561e7a413e30ba", "generation_seconds_batch": 22.865598909091204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3089518547058105, "target_z": 1.8828125576380494, "ce": 6.022613525390625, "genuineness": 4.18099308013916, "identity": 0.7289091348648071, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ca874c1dabf0677868e2b5e85a094aff611bde6a427fef50a32eacca6fdc5f80.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "cabd0b5f1ca029bb0ed1b6b6acc327337f1133f717810f46b9dfcb75a63e2ce7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "cbbd40e7b23b6e70d80e618dc9e5f6b51395618c51437b911c3d759e15bb7bc2", "generation_seconds_batch": 22.370217808056623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.3641486167907715, "genuineness": 4.557555198669434, "identity": 0.7807872295379639, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cabd0b5f1ca029bb0ed1b6b6acc327337f1133f717810f46b9dfcb75a63e2ce7.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "cb84d8bd29980e6a223cef9af08717cc7a75f975d3f63a6b2c027a77d0a30fbf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b5b864030b3f0e4b0574211ce1b97e0e339924ca7c6a877f694136cfb7735fa2", "generation_seconds_batch": 24.070775900036097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.268836498260498, "genuineness": 4.20609712600708, "identity": 0.5830391049385071, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cb84d8bd29980e6a223cef9af08717cc7a75f975d3f63a6b2c027a77d0a30fbf.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "cbcdf1df9c62996087999bc82226331dcdb1b3f78d0f0b6321fc71a10ec10913", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "604a9b59751f613e1470fe1a71e5bff01cf696a5df5eac6c2401c09f2ee53f1b", "generation_seconds_batch": 21.91388225799892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941354513168335, "target_z": 1.4765624588299646, "ce": 6.070269584655762, "genuineness": 4.632867813110352, "identity": 0.738823413848877, "hypothesis": "Je trouvai 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_WARM_reg_high/cbcdf1df9c62996087999bc82226331dcdb1b3f78d0f0b6321fc71a10ec10913.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "cc4e6844ab5c507e26004bd3c37755aeca3bdc6d45f0dab477ae2369c9bbef32", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "92a20cd36e8fc94a12a163e25d8bdd5568cfaf96d6ba56b9bcdea4c8fde785e9", "generation_seconds_batch": 23.794667519978248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5384113788604736, "target_z": 1.0312499670639719, "ce": 6.109982967376709, "genuineness": 4.15588903427124, "identity": 0.7780380249023438, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cc4e6844ab5c507e26004bd3c37755aeca3bdc6d45f0dab477ae2369c9bbef32.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "cd4602e4f0390727fdd567825efc6b4e07112944d3fa9460b64b4f60f17fa901", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f3b12e101661f42fa30ba9d9ba672a362833f4d96489731e98ccaf9b27ea522e", "generation_seconds_batch": 16.32790992502123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.395919322967529, "genuineness": 4.18099308013916, "identity": 0.5114370584487915, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cd4602e4f0390727fdd567825efc6b4e07112944d3fa9460b64b4f60f17fa901.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "cd82101f56f184cd6efaddf35e5ea25606ecadff082647e30f19fe6091173aac", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ead2f070ff0b42cb59a967ee30d414121d8edb855ba463db59ebc54eb16109e8", "generation_seconds_batch": 25.01297877705656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.47632867097854614, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cd82101f56f184cd6efaddf35e5ea25606ecadff082647e30f19fe6091173aac.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "cd846bc0e00bf37246f0abad9c8ea94369e6db02ced53bc2029934e7eb0c413b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "e1e91f3125f30b018c37b85f5ff25497f7931915ce2fc62866e54749dae20366", "generation_seconds_batch": 21.432926325011067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.040323257446289, "target_z": 1.5859377387862046, "ce": 6.24500846862793, "genuineness": 3.898571252822876, "identity": 0.29606103897094727, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cd846bc0e00bf37246f0abad9c8ea94369e6db02ced53bc2029934e7eb0c413b.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4f6677cc099cf2370597c6e083d0fcdc7ca304399d99eaf162435bf8a94874f4", "generation_seconds_batch": 16.645318311988376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9554929733276367, "target_z": 1.4921875741060635, "ce": 6.4515180587768555, "genuineness": 3.9927117824554443, "identity": 0.2985881567001343, "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_WARM_reg_high/cd9417d7bfaaf4c9eb39b90fe829dd381452cc047777c14131a94331ffc48b20.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "cd96024a2d14a4b6da835d9662e9e4369fe1d9e811df0908818d4f21ba2098de", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "afd2b878dc6ad0946c4fa0c2a61705efd38246a1e716030654e110368aa104a3", "generation_seconds_batch": 16.364639181061648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.109982967376709, "genuineness": 4.20609712600708, "identity": 0.7361953854560852, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cd96024a2d14a4b6da835d9662e9e4369fe1d9e811df0908818d4f21ba2098de.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "cda18039715acbf3975fa448f132e0e0e7da4865e1b78420a852b98e0263cc88", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3b0e5634468dc618b6fd1b87db1c4671eaa93bd38dfef241279ca12a836eab67", "generation_seconds_batch": 23.849725817912258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.2767791748046875, "genuineness": 4.231201171875, "identity": 0.5976148247718811, "hypothesis": "Wat is? Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.16666666666666666, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cda18039715acbf3975fa448f132e0e0e7da4865e1b78420a852b98e0263cc88.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "cdd04c76ee0e584ca2280782bba6b65b058284b9ba019fe262cc9faaf43830ff", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "bb33655be6fc4784cb14ca73247bf863a756726e8ea6ad832f75d817910e02b2", "generation_seconds_batch": 22.68763935903553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.365505218505859, "target_z": 1.945312491765993, "ce": 6.09409761428833, "genuineness": 4.231201171875, "identity": 0.6563866138458252, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hacía.", "wer": 0.11764705882352941, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cdd04c76ee0e584ca2280782bba6b65b058284b9ba019fe262cc9faaf43830ff.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "cdebd9b1608c240043aba91407a664722b792ae9c622135641dc423f4b972423", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "283d4f36fd2dfa94821fe13a9f5117a122dfbaf3176257d1bd638564102ab7ef", "generation_seconds_batch": 23.026826991001144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.1020402908325195, "genuineness": 3.973883628845215, "identity": 0.4391959309577942, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cdebd9b1608c240043aba91407a664722b792ae9c622135641dc423f4b972423.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "cdec75e7fc3a9b22feb4fbf5ed8181712b3f4ee7df57a1dd5c80eb67e3ddf337", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "472df367a48090dcd234e86e2867da5eede5bbfee9052772ec817f7b3ea90ace", "generation_seconds_batch": 27.55826400592923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5596189498901367, "target_z": 1.054687508234007, "ce": 6.260893821716309, "genuineness": 4.017816066741943, "identity": 0.3202736973762512, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cdec75e7fc3a9b22feb4fbf5ed8181712b3f4ee7df57a1dd5c80eb67e3ddf337.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "ce1563d38f80a50a7a860a3532482f46d61ac2b8d7f2d1fe1ef02c99915d9298", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a9792dc7aa734f2f965ee950c00c9cdfc66a89661fc956677c8d7d80e1133bbd", "generation_seconds_batch": 18.43187206494622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.238260269165039, "target_z": 1.8046877717222327, "ce": 6.197352409362793, "genuineness": 4.193545341491699, "identity": 0.8082505464553833, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ce1563d38f80a50a7a860a3532482f46d61ac2b8d7f2d1fe1ef02c99915d9298.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "cebcf22d27450c3dac718a11054db4fb98cc00e8acf7469b55ad56eba09efc81", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4bed3b988314851af81d064cef28db5d42197719e797eba5fd6b90e9c3cb048c", "generation_seconds_batch": 22.918663172982633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4641849994659424, "target_z": 0.9492187047129612, "ce": 5.95112943649292, "genuineness": 4.256305694580078, "identity": 0.7704318761825562, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cebcf22d27450c3dac718a11054db4fb98cc00e8acf7469b55ad56eba09efc81.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "cf06e211208bc056907663cfdb2b06f49b871597be03ee2a85f7c8d55fd70598", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0ee5ee930eafd253242bdc0f802d8d97287945f71528531a714e1c0edc80359f", "generation_seconds_batch": 16.781978926970623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.205295085906982, "genuineness": 4.3127899169921875, "identity": 0.6307692527770996, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cf06e211208bc056907663cfdb2b06f49b871597be03ee2a85f7c8d55fd70598.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "cf0d52860543a4f0b0f1a872ecc54a697f8d10d2e46651a4ab6aab89ec3dee2e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a60901a83bb8e03a658a5d1d1e53de3aff9d3045cea8c9a2743b1ef5faabb3a0", "generation_seconds_batch": 16.32337344996631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.300607204437256, "genuineness": 4.306513786315918, "identity": 0.695290744304657, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cf0d52860543a4f0b0f1a872ecc54a697f8d10d2e46651a4ab6aab89ec3dee2e.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "cf1914531df03e252ddd366a7195718f856f99de1bd84d0bb009673bcb7266d8", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "abe70767df291ccdd03a1de4041c7fc33073a0fd3e28c7b72e563446232c4aa1", "generation_seconds_batch": 17.405034702969715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.070269584655762, "genuineness": 4.3504462242126465, "identity": 0.5321551561355591, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/cf1914531df03e252ddd366a7195718f856f99de1bd84d0bb009673bcb7266d8.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "cf3d81d824f1a9613564e33d0274d0e4e7b86ea87cd15b7da894d07630f46977", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4e6be67cd4b453ef9fee38c4bcb342be2dce75f30815d3948012084a591df080", "generation_seconds_batch": 22.107812745962292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3510782718658447, "target_z": 0.8242188364570741, "ce": 6.316492557525635, "genuineness": 4.319066047668457, "identity": 0.7287138104438782, "hypothesis": "De deur 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_WARM_reg_high/cf3d81d824f1a9613564e33d0274d0e4e7b86ea87cd15b7da894d07630f46977.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very warm vocal delivery. 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": "7f612a481b90841b5866468da9301d5627c4da4d3fe0c8f0b5f58ef780321492", "generation_seconds_batch": 19.759389695012942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4606504440307617, "target_z": 0.945312491765993, "ce": 6.3641486167907715, "genuineness": 4.3504462242126465, "identity": 0.4946610629558563, "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_WARM_reg_high/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "d075cc67527787b43f3606dd0dcee11b685a9f15b967ba078836f771054918ac", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "14cac198902e0be7eed2df464c039a78b342a4e5e14b1b3e591e4a673e05300b", "generation_seconds_batch": 15.588876973022707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.515059471130371, "genuineness": 4.300237655639648, "identity": 0.542892336845398, "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_WARM_reg_high/d075cc67527787b43f3606dd0dcee11b685a9f15b967ba078836f771054918ac.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d09d2f6b7eb0118269f4728a61f992dd6220bfeb01061eb67be8e9464f4332dc", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0463e3e446110df840744ab92d4d9880f9b2d6463abd82271e3d6a758ed7fc44", "generation_seconds_batch": 21.281110243988223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.202914237976074, "target_z": 1.7656251152760987, "ce": 6.006728172302246, "genuineness": 4.2751336097717285, "identity": 0.541541337966919, "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_WARM_reg_high/d09d2f6b7eb0118269f4728a61f992dd6220bfeb01061eb67be8e9464f4332dc.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "d0cebfb635b30241d63e17f08c686f146c80e1e7d685861fa8fd5cafce702ef1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7e66ea8131c0c2e9a81fa7fdcbf154423a3f011d683068df6b7cbbd18c9b9047", "generation_seconds_batch": 21.854197255917825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.125868320465088, "genuineness": 3.9299514293670654, "identity": 0.600949227809906, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d0cebfb635b30241d63e17f08c686f146c80e1e7d685861fa8fd5cafce702ef1.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "d0f72c8959c26623f573c4b58a23788e32e898db2bb8c319d56213df916451d9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ad8031770fe8e5e67e3a68d6cf3a170d057daf8aee1b391d904cbe09ee7b1ce9", "generation_seconds_batch": 17.919459426077083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 6.197352409362793, "genuineness": 4.256305694580078, "identity": 0.8263294696807861, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d0f72c8959c26623f573c4b58a23788e32e898db2bb8c319d56213df916451d9.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(very warm vocal delivery. 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": "20ec4f72ec4f107510956a134cce7790af7913a1064752c6946e87b83d80782b", "generation_seconds_batch": 18.64826293999795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686864137649536, "target_z": 1.195312491765993, "ce": 6.300607204437256, "genuineness": 4.080576419830322, "identity": 0.44981849193573, "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_WARM_reg_high/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d1428e645822c91ee6d999c9063009706e091d39c96ed9564ea33e91c7202cf0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c9f54856b4d129f66045087905b6940cf8fdbffe18a02a555e44fc42e89ffec4", "generation_seconds_batch": 20.473791806958616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.054461479187012, "target_z": 1.6015625905740776, "ce": 6.0146708488464355, "genuineness": 4.419482707977295, "identity": 0.7323325872421265, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la casa.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d1428e645822c91ee6d999c9063009706e091d39c96ed9564ea33e91c7202cf0.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "d14a3ac6efdc99e9f7653bc095aa8d453638e516798ada30d24c298408238ec3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2752bd367e14e53a7f7d5317c25fdd9e22623baa036376465993b58bded2f018", "generation_seconds_batch": 21.929223209037445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.173524379730225, "genuineness": 3.8169827461242676, "identity": 0.5351298451423645, "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_WARM_reg_high/d14a3ac6efdc99e9f7653bc095aa8d453638e516798ada30d24c298408238ec3.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "53cec0f2a0a3f90ec69f6885dac59d60d3497bfd84f5fa09a587ca3e840966da", "generation_seconds_batch": 16.95845624397043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2556443214416504, "target_z": 0.7187500329360282, "ce": 6.300607204437256, "genuineness": 4.218649387359619, "identity": 0.34391745924949646, "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_WARM_reg_high/d199fa9cf88e674546620dee75a036ad19adde547e4a340d1c7badf3607578d7.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "d1b74fe5135e79417a2cfa7b53a0960709e87a9f0e9538a8f9c8ed23f7e68212", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "7641b9ae07d321ec1c19651e0a52e93c3eba369405527da4f3dc83cb7e2765f7", "generation_seconds_batch": 23.435381450108252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.068599700927734, "target_z": 1.6171874423619506, "ce": 5.8875885009765625, "genuineness": 4.65797233581543, "identity": 0.7433143258094788, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d1b74fe5135e79417a2cfa7b53a0960709e87a9f0e9538a8f9c8ed23f7e68212.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "d1ecd59e7b4d4183d2cec953b19bdc159c0458c176645dd2f7da7086e2a8a6b5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "3c1f314dcabd3c72ee89a3283f7c938daf240c4bde7c4824062ba228dab8258e", "generation_seconds_batch": 23.771759801078588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0332536697387695, "target_z": 1.5781247859158165, "ce": 6.006728172302246, "genuineness": 4.482243061065674, "identity": 0.7158330678939819, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d1ecd59e7b4d4183d2cec953b19bdc159c0458c176645dd2f7da7086e2a8a6b5.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "d200cd5c92490bd970ed4575885fc5d1a011c06d218e456cb54ab8cdf3db3fea", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3c1d7a60bc9ce67183a839b43374acc4b40ca588ead79c19f73ea5da1b684fec", "generation_seconds_batch": 22.676408263971098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.252951145172119, "genuineness": 4.306513786315918, "identity": 0.6883071660995483, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d200cd5c92490bd970ed4575885fc5d1a011c06d218e456cb54ab8cdf3db3fea.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "d207337648cc06062f0b867ffef6879307c14a40f70fb0f4fe1e0c9ea8382637", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c1d2231f99f70e08089b345e1d165a7d97c89c532685d40f9c8e93c3666273f6", "generation_seconds_batch": 20.43076894094702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3758203983306885, "target_z": 0.8515625905740776, "ce": 6.308549880981445, "genuineness": 4.507347106933594, "identity": 0.445634126663208, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d207337648cc06062f0b867ffef6879307c14a40f70fb0f4fe1e0c9ea8382637.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(very warm vocal delivery. 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": "6a8a281fa300f3fbe7beb1dd61330d13db9ea0291378dd3bb7e1d3820f5fc795", "generation_seconds_batch": 17.594193930039182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 6.23706579208374, "genuineness": 4.432034492492676, "identity": 0.7255384922027588, "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_WARM_reg_high/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "d2ccadfa93c3d9277acbd118da8048cb81b4c105b1c50f901968e128da372136", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "aa64531b4311b2ca04bb1899d94a0d2516b18fe5edf0b426088275cf6b6b65c8", "generation_seconds_batch": 29.340077362023294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.340320587158203, "genuineness": 3.9676074981689453, "identity": 0.4865025579929352, "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_WARM_reg_high/d2ccadfa93c3d9277acbd118da8048cb81b4c105b1c50f901968e128da372136.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d3a9e9ba16f61f51097cf89f415c79b44e1e730d201368b7df5b5a9563e319f7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "670e821feff2c1c4d4eff8661061b95d0df0c991a8427c2d0e288c35a189da84", "generation_seconds_batch": 24.519808424985968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.09409761428833, "genuineness": 4.2249250411987305, "identity": 0.4404034912586212, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d3a9e9ba16f61f51097cf89f415c79b44e1e730d201368b7df5b5a9563e319f7.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(very warm vocal delivery. 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": "aea1af51b7fa6682bc9437159d0fec5842389840fdd462508e8c6dc39fb75b1f", "generation_seconds_batch": 15.778318897006102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.054384231567383, "genuineness": 4.482243061065674, "identity": 0.653449296951294, "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_WARM_reg_high/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "d400c801120cf7e762ce2db5ef25f58b49872951db80a0b950f9d0d7d904909e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "73906da61d381eb73d08126d3eb899aaf2ca5b7b4a974a50088b4fe4c0760ac1", "generation_seconds_batch": 21.91388225799892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 5.895530700683594, "genuineness": 4.720732688903809, "identity": 0.7714831233024597, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d400c801120cf7e762ce2db5ef25f58b49872951db80a0b950f9d0d7d904909e.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "d409ca24acfc1011bedb44c3ed4146ea29cbc072d4bb347924d207500b86426d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7dd47c026b27fd30623cbea39c07d4ca2577bcde05b702af9fe0dbad91ca9eec", "generation_seconds_batch": 23.73602718207985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.308549880981445, "genuineness": 4.168440818786621, "identity": 0.576555073261261, "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_WARM_reg_high/d409ca24acfc1011bedb44c3ed4146ea29cbc072d4bb347924d207500b86426d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d450921be499b0a8fec99fafb3af648dfec3d5b0af37aacd7dc033d356ff1615", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c3202a5568026875388ee58b7b6682b0babc0995a067373bfb673b8c7d87fad5", "generation_seconds_batch": 24.805956483003683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.125868320465088, "genuineness": 4.394378662109375, "identity": 0.7821095585823059, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d450921be499b0a8fec99fafb3af648dfec3d5b0af37aacd7dc033d356ff1615.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "d45bacdb6b120f67ba69a9c805e3e1aa5527a7a7dc35c8be74666b1903691b6b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "92825928e765a016333f690ddcba8ada1bb258f613747bfc23bfc3358be4e214", "generation_seconds_batch": 23.654634250095114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.435632705688477, "genuineness": 4.331617832183838, "identity": 0.44038575887680054, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d45bacdb6b120f67ba69a9c805e3e1aa5527a7a7dc35c8be74666b1903691b6b.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "d48e563e0bbcd19c9f41bb31e3c4e7df001b74bbe2f02762fd61956903c6442e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "6476f746dd20ca9d3dc84a3fc158b1548b4de3bc65d87a348fc912392ebefe2d", "generation_seconds_batch": 28.663059689919464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.316492557525635, "genuineness": 4.557555198669434, "identity": 0.7177350521087646, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d48e563e0bbcd19c9f41bb31e3c4e7df001b74bbe2f02762fd61956903c6442e.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "d4c8035973a9a2971bbcbd4be2ec442781f6ddc922cd688505da833f74ed5b71", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "adeb3bf40b5ed677e3fccb1552fcee7ffb15158a84932cb771177c06a4debc33", "generation_seconds_batch": 24.955271175014786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.259467601776123, "target_z": 1.8281250494040424, "ce": 6.356205940246582, "genuineness": 4.293961524963379, "identity": 0.48684418201446533, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d4c8035973a9a2971bbcbd4be2ec442781f6ddc922cd688505da833f74ed5b71.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "d4e2356120543e8689295dfb6204e3306c59d89d0ffe6be91bd6d4067d430dc2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "0d61075526a7d1838c6c9ea846998c3284c577ab0b9207ff4bdca31122713eb6", "generation_seconds_batch": 16.229844363057055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7716941833496094, "target_z": 1.2890623929579084, "ce": 6.165581703186035, "genuineness": 4.268857479095459, "identity": 0.25635048747062683, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d4e2356120543e8689295dfb6204e3306c59d89d0ffe6be91bd6d4067d430dc2.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d4f7cb5a296d9d5aec0b77d3e08d7128c6603635c97e015ac454e20eb6b14112", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "84f6c2cc87bf7686911888b5b575b70df6062afa4af53610cb86d4cdb70ce6b5", "generation_seconds_batch": 26.567971289041452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.6427612900733948, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d4f7cb5a296d9d5aec0b77d3e08d7128c6603635c97e015ac454e20eb6b14112.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "d523f19490379336bb143c4a101364d93bf0cd75f43cf9188f722217b642b7df", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6433f809ff0ba8f3b401537d58ab7f418dce4d18301a1d948808ce21ad875e85", "generation_seconds_batch": 16.143981837900355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5066001415252686, "target_z": 0.9960937870530318, "ce": 6.332377910614014, "genuineness": 4.080576419830322, "identity": 0.5225043892860413, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d523f19490379336bb143c4a101364d93bf0cd75f43cf9188f722217b642b7df.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "d58fcecf96be40f449408e558a2613e8b056cf1323be84265d23f4ae4906e56d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4485e4a53e845d202afd8620e1f2770536104c39fa6afec22ec2f1580740f10a", "generation_seconds_batch": 20.275813754997216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.799971103668213, "target_z": 1.320312623510106, "ce": 6.443574905395508, "genuineness": 4.356721878051758, "identity": 0.277915894985199, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d58fcecf96be40f449408e558a2613e8b056cf1323be84265d23f4ae4906e56d.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "d60582b0439c27b2424cf6460d0833480f7347a0f565df00065968e3b6d4c8ce", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5def4c295d49a2a5b4c37e198b2366bfced8c0aed9f63e7a07bf6400833a70e1", "generation_seconds_batch": 24.336334924097173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 5.911416053771973, "genuineness": 3.936227321624756, "identity": 0.6726138591766357, "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_WARM_reg_high/d60582b0439c27b2424cf6460d0833480f7347a0f565df00065968e3b6d4c8ce.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "d6d02392a3c1e917f38de2aaaff321c84be3596405d172c577be53a66aae9017", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fb2fdd9be519e47f19a3a0c23f98c3c57ad9561c3286416dc78402fe6eb44c19", "generation_seconds_batch": 25.807430564891547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.252951145172119, "genuineness": 4.143336772918701, "identity": 0.26455965638160706, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d6d02392a3c1e917f38de2aaaff321c84be3596405d172c577be53a66aae9017.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "d75108e3c23dd0d25130a0af21952988bf3381e647412e6b687627c18f954e51", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d898d0a7b758151758e41bf438791542720a64a8743cea8e53bc5d2b9c91feba", "generation_seconds_batch": 21.756471942993812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.268836498260498, "genuineness": 4.331617832183838, "identity": 0.7091675400733948, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d75108e3c23dd0d25130a0af21952988bf3381e647412e6b687627c18f954e51.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1352136cd78eea25867a61648cc02c511861117036a38f7f4004234e869730f1", "generation_seconds_batch": 23.503982845926657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.665656566619873, "target_z": 1.1718749505959576, "ce": 6.3641486167907715, "genuineness": 4.068024158477783, "identity": 0.6797455549240112, "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_WARM_reg_high/d751aa98fda13e503fac763c598a3e48f410bf2556d324e99048de2a91897b70.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "d7669b9e203b08dce4253c8ccdd053d807151b0418f3ed4fc02cd1c2f572776e", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5be3b3c6ef36353710c5781e25c376bcef22e48c6eb1b397a8e4becbb772f740", "generation_seconds_batch": 16.04885371995624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 6.38797664642334, "genuineness": 4.645420074462891, "identity": 0.6680570840835571, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d7669b9e203b08dce4253c8ccdd053d807151b0418f3ed4fc02cd1c2f572776e.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "d7836aa0d268dc0fad81a25c8cc324eade4e53d539925ab6afe96f1ddd22a7e8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "06e4ced9ea772b4ab0bb2510332678feefa4f32aaed8672f8c5e9b769bfefd53", "generation_seconds_batch": 17.35024698509369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.252951145172119, "genuineness": 4.262581825256348, "identity": 0.712303638458252, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d7836aa0d268dc0fad81a25c8cc324eade4e53d539925ab6afe96f1ddd22a7e8.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "d7feeb1b08fcc4067f8546b1fad72c7754befc0ba5be5d21ae6ab5210c208abf", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b1abf79718aa56d9aa68c222c6335c7522ec43ee47b48acc6031633d1c20b2d6", "generation_seconds_batch": 26.50251111097168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898939609527588, "target_z": 1.42968763997812, "ce": 6.2767791748046875, "genuineness": 4.494794845581055, "identity": 0.6660511493682861, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d7feeb1b08fcc4067f8546b1fad72c7754befc0ba5be5d21ae6ab5210c208abf.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "d82076ce85e83e7428d083fafa1d4ac17ff5ec654b10642d2c215b4e2c51aa6f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8d4235f719ae812acab1ab1f14e9a5b6c3b9b8c893cd53fc9a04d649b1624d48", "generation_seconds_batch": 23.426912466995418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.157639026641846, "genuineness": 4.105680465698242, "identity": 0.5652991533279419, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d82076ce85e83e7428d083fafa1d4ac17ff5ec654b10642d2c215b4e2c51aa6f.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "d83f0b1542e588f31c06232b9c3a18b8fce587df8ec8fea6515605653bca6a58", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "71741bc0a855f9b09405ff4a40b8fcddbd209208bf152e7dd737598390b75aa1", "generation_seconds_batch": 22.816591658978723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.157639026641846, "genuineness": 4.337893962860107, "identity": 0.5358686447143555, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d83f0b1542e588f31c06232b9c3a18b8fce587df8ec8fea6515605653bca6a58.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "d84473dfb8db86b5d995bf85abb5b0a7b3c1aa39ee8edc23d727728f6b8ace98", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "631ec8cda4fcd899531584942f02a7da9a0a370cf6d6b7bd60da30b9bd5cd660", "generation_seconds_batch": 16.961068186094053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.372285842895508, "target_z": 0.8476563776271093, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.5499560832977295, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d84473dfb8db86b5d995bf85abb5b0a7b3c1aa39ee8edc23d727728f6b8ace98.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d8c6e17e3901735014c68b277a5401c6bb551251b896be8f20b86351a3482805", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a8dd804cd1c9e96864b4eed9e5691e85b54a3e05bca483b5e7e51481a52ce3c3", "generation_seconds_batch": 20.473791806958616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.5513311624526978, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d8c6e17e3901735014c68b277a5401c6bb551251b896be8f20b86351a3482805.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "d935dba93f122213c337ab313dd692bf7bccc5cea62505ae3e55b9494c2de542", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "eaa3ba49f09a9f8b583b934f2a265121120e8c990c6b749db67308695ea4546a", "generation_seconds_batch": 16.910009634913877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3404743671417236, "target_z": 0.8124999341279435, "ce": 6.332377910614014, "genuineness": 4.457139015197754, "identity": 0.4562622308731079, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d935dba93f122213c337ab313dd692bf7bccc5cea62505ae3e55b9494c2de542.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "d93da742e7ef84ee583b731fdf8cc5f80a2839abb52614aeca115670e651a54e", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "df5200407c48e2c490305ac32cfdaa4b146d7746fb31e5ad2100de4eb425ea71", "generation_seconds_batch": 17.229789471020922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.141753673553467, "genuineness": 4.193545341491699, "identity": 0.6573449969291687, "hypothesis": "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_WARM_reg_high/d93da742e7ef84ee583b731fdf8cc5f80a2839abb52614aeca115670e651a54e.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0df92172d0e1160d0a750eb02ebc23fa7a7a9a1e14811f836c7b40b885fdfc98", "generation_seconds_batch": 21.47452757600695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 6.054384231567383, "genuineness": 4.193545341491699, "identity": 0.6877723932266235, "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_WARM_reg_high/d9703bcb71f09fa3ae68f156919fafd90507f99ddbdd8c71c60e1eeb60e29e68.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "d9b49451d5062a88816fa9cf13f9005d08b332301ab89ff16fb51735a377b2f5", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "512ef30a969e9d0ea7bed9e2f181f08c6db139fdf62e8ed79837586cb81c67c7", "generation_seconds_batch": 15.254763682023622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3616819381713867, "target_z": 0.8359374752979788, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.866986095905304, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d9b49451d5062a88816fa9cf13f9005d08b332301ab89ff16fb51735a377b2f5.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "d9edee207865118b171ef6b346264925d8138d61eb0a32abe539deb25bdd476d", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "455878fc60e409743c8a7a179babbabc8c5c68c12db4ca66a192f24cf4da8c8f", "generation_seconds_batch": 17.35467329097446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.507116317749023, "genuineness": 4.570107460021973, "identity": 0.2761961817741394, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/d9edee207865118b171ef6b346264925d8138d61eb0a32abe539deb25bdd476d.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "da56684f568f8800a05d0387e970a8566474f357d6214104d1ec05adaa4a1575", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "c81b74747a4b1f2c1c099e03f8c4191a378e36c56f9ed20c92250d31f51a653c", "generation_seconds_batch": 24.12149797496386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.38797664642334, "genuineness": 3.8295347690582275, "identity": 0.1989797055721283, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/da56684f568f8800a05d0387e970a8566474f357d6214104d1ec05adaa4a1575.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "da6c060f9f87d3d67fad6e4ce1924ed0e1e34403243b8c46b8c264eac6621e40", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "87c5a825d4ce872952791deafb0bd5363066f6588de03c22aa3518c9af9d65a4", "generation_seconds_batch": 20.9369902969338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.252951145172119, "genuineness": 3.760498285293579, "identity": 0.396364688873291, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/da6c060f9f87d3d67fad6e4ce1924ed0e1e34403243b8c46b8c264eac6621e40.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "da955429a733b11ea31ee455e85b4ef934aea789e1decc97544af26fec5f9b14", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "553c7312fb5ca2555c0598d268c613af690823d2eeb1886f0e025926efc09025", "generation_seconds_batch": 16.81424726091791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.348263263702393, "genuineness": 4.394378662109375, "identity": 0.29446813464164734, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/da955429a733b11ea31ee455e85b4ef934aea789e1decc97544af26fec5f9b14.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "da96bc988603a228751b330df37d3039a5ac15268e810b008fbf1eb4abf28586", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "087dd7b75a37b88b81f86beefb96f62157cb652a78022f9eba7c5e85e5c45d56", "generation_seconds_batch": 22.865598909091204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344297409057617, "target_z": 1.921874687107732, "ce": 5.990842819213867, "genuineness": 4.494794845581055, "identity": 0.789771318435669, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/da96bc988603a228751b330df37d3039a5ac15268e810b008fbf1eb4abf28586.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "da9962fa526f33b9f71be2052062a93ffe6b6310b10c1dff72ba581c16fe72ec", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "c5adc5a8b07c19feb0d27b8b2ca41e112852f54967964b2295cb20fcdfcc176f", "generation_seconds_batch": 20.9369902969338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.284721851348877, "genuineness": 3.9048471450805664, "identity": 0.34797197580337524, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/da9962fa526f33b9f71be2052062a93ffe6b6310b10c1dff72ba581c16fe72ec.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "daa3931826107e4b6f5f84b42f3533a467b3c1005ff3cddad3d23f2a80e899f7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "76bf283d3dc46dab66533d7b45e267d43edf9cdf55f7c4028cd60e394729021d", "generation_seconds_batch": 18.204944938072003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.545480728149414, "target_z": 1.039062656446134, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.7424144744873047, "hypothesis": "De deur 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_WARM_reg_high/daa3931826107e4b6f5f84b42f3533a467b3c1005ff3cddad3d23f2a80e899f7.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(very warm vocal delivery. 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": "0c0afb6cdc0f4ff3e4e3ae6f7212835b20ab57bb8343ea36b273659f4f2aec30", "generation_seconds_batch": 19.319175221957266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.478323459625244, "target_z": 0.96484381998906, "ce": 6.475345611572266, "genuineness": 4.250029563903809, "identity": 0.2248324304819107, "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_WARM_reg_high/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "db94b472d6e675853525c14945efaa0f3a8f1ee98fa73c67f84a67b924ffc9e7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6c1d2746c4da2e75dd306eed09f669766c131df2ada98149375be7231942929b", "generation_seconds_batch": 25.097747804014944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9554929733276367, "target_z": 1.4921875741060635, "ce": 6.332377910614014, "genuineness": 3.9676074981689453, "identity": 0.5589628219604492, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/db94b472d6e675853525c14945efaa0f3a8f1ee98fa73c67f84a67b924ffc9e7.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "dc28725ded6b367d550996b6974f2d7a3262a55fee402ea17382d225edf113a8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ef9abed1612cdde7e95816f886f56e159490bc1578f11185215a55e3c4f98516", "generation_seconds_batch": 23.649388913996518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 5.998785495758057, "genuineness": 3.7793264389038086, "identity": 0.6459268927574158, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/dc28725ded6b367d550996b6974f2d7a3262a55fee402ea17382d225edf113a8.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "dcaad17153993dbb4c0f12e4a997554e21cf703c8391d9f260650e96b23851b9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0335818a95cfac66b4a062cb9ef284465a89ebe6b8e7e6ee361d88ef7a60c1aa", "generation_seconds_batch": 17.94504861102905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.324435234069824, "genuineness": 4.243753433227539, "identity": 0.35977113246917725, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/dcaad17153993dbb4c0f12e4a997554e21cf703c8391d9f260650e96b23851b9.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "dcbab1396612fa505bd599904d2838697f0c812c1674e558459351a8fff4024c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "488c6781f23b6984da77621ad8e7312e6962e40603dedc842e7f2deb469bf9b7", "generation_seconds_batch": 17.076868982985616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.351366996765137, "target_z": 1.92968763997812, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.7075049877166748, "hypothesis": "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_WARM_reg_high/dcbab1396612fa505bd599904d2838697f0c812c1674e558459351a8fff4024c.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "79d99885cd5d4177af9224f3d46e8c16bef86f1c4a36d8f37c4f09dc940d5645", "generation_seconds_batch": 21.360847497940995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.475345611572266, "genuineness": 4.419482707977295, "identity": 0.30811309814453125, "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_WARM_reg_high/de1483167c8fe2ba6c4adeb7f904bb635d1b5e80346ee62f36c47eb034ddd8bd.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "de56da14e4c290035bd6f6feb618b82ad50b7e3f3dbb91fedc4585f5a5c50a0b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a546a453dbc3965a23f8c16e34ba6c42671a04056736ffa4aa192c68e336c22e", "generation_seconds_batch": 23.64911935303826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.117925643920898, "genuineness": 4.042920112609863, "identity": 0.5330542325973511, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/de56da14e4c290035bd6f6feb618b82ad50b7e3f3dbb91fedc4585f5a5c50a0b.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "de699e418ca2a8b5cfbcba3b033f0376110bb9c05d35715e7cf8b14c3b5761db", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a10f8ff66398aca8df1e2e0dbc4814e2ede752096ec39e43a72c61a93040cb5b", "generation_seconds_batch": 23.797212996985763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.828247547149658, "target_z": 1.3515623270858519, "ce": 5.959072113037109, "genuineness": 3.798154592514038, "identity": 0.5125914812088013, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/de699e418ca2a8b5cfbcba3b033f0376110bb9c05d35715e7cf8b14c3b5761db.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "decec7c0479ded03884219d47756f5eafc5bd446ab71d39f60c05d415efe57bd", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "099ba680e7872209b0f4445ca61fa53530c983dc383189e6caf597f44060c626", "generation_seconds_batch": 22.840937337023206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.884801149368286, "target_z": 1.414062524702021, "ce": 6.372091293334961, "genuineness": 4.030367851257324, "identity": 0.6086670160293579, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/decec7c0479ded03884219d47756f5eafc5bd446ab71d39f60c05d415efe57bd.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "df031e86144e795ebadcb21554dd917183a8f3b1ef3c7decf5a6a5b8e121ffa5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4f5c274d1d550e1eb5b512b277c1f81466502f5ff5f6f0a6f86a38d10b11aba8", "generation_seconds_batch": 15.446155859041028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.062326908111572, "genuineness": 4.319066047668457, "identity": 0.6758586168289185, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je kleid.", "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_WARM_reg_high/df031e86144e795ebadcb21554dd917183a8f3b1ef3c7decf5a6a5b8e121ffa5.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "e007967c33a3dc9a53b376c846df4965f51cbfe4cbde5bbb98a47fffb822311d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a0eabc2bba5e00252c2f957f597861af178024889e75c1d2aa3e1bb13951c952", "generation_seconds_batch": 17.409799593966454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.217052459716797, "target_z": 1.7812499670639719, "ce": 5.982900142669678, "genuineness": 4.545003414154053, "identity": 0.7053890228271484, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e007967c33a3dc9a53b376c846df4965f51cbfe4cbde5bbb98a47fffb822311d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "e0c1ea55461512dfd7435965b4539646418d3a3d4b8c89b484344609a00852a1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "753b195900de4addf48b7ddedef39927f4b6e0f0103502d59a5806e352ce7256", "generation_seconds_batch": 24.134065633057617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5280961990356445, "target_z": 2.124999868255887, "ce": 6.1020402908325195, "genuineness": 4.457139015197754, "identity": 0.6276500225067139, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e0c1ea55461512dfd7435965b4539646418d3a3d4b8c89b484344609a00852a1.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "e114fe59588041c93a51f3ffea1640e107d1930e2ad0057aeca8b04eb797a925", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "04e5d0906bc92d108afb6e788b18b271d77cbdcaf098eb8a5d62c2f3f2b09113", "generation_seconds_batch": 16.60803925804794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.634199142456055, "genuineness": 4.2374773025512695, "identity": 0.7342280745506287, "hypothesis": "Ik 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_WARM_reg_high/e114fe59588041c93a51f3ffea1640e107d1930e2ad0057aeca8b04eb797a925.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "e2306474cb7a6d29cc50ccabf100c9a21d8742f9b496cd5604a6fe116ee93da1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "df7c502636d3787dbe1e9be2c91b769de3f0178175c7c8648952ec2f60988030", "generation_seconds_batch": 17.193626957014203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.153429985046387, "target_z": 1.7109376070420916, "ce": 6.221180438995361, "genuineness": 4.080576419830322, "identity": 0.11611509323120117, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e2306474cb7a6d29cc50ccabf100c9a21d8742f9b496cd5604a6fe116ee93da1.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "e2ad12bc442a231ce3d92d595b9a11600ae82001eb3487233da7b583c734cec1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d22c07018aa6248b3ffa2ca0850cc28f983ea19eeb2274ceebcd2f93567b325e", "generation_seconds_batch": 17.764955307007767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686864137649536, "target_z": 1.195312491765993, "ce": 6.205295085906982, "genuineness": 4.444586753845215, "identity": 0.45350050926208496, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e2ad12bc442a231ce3d92d595b9a11600ae82001eb3487233da7b583c734cec1.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "e30555747f841c7a786c08c094774295e511f411160625235f9ea044727c4262", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "24272bea419b4df1fd2292cf1a73650eb771efdad298cf15078531cdf0f614b8", "generation_seconds_batch": 25.920054264948703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.157639026641846, "genuineness": 4.20609712600708, "identity": 0.6819846630096436, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e30555747f841c7a786c08c094774295e511f411160625235f9ea044727c4262.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a4a44bc4a4d5bf81973f729ff27a6a36badcd46720384fc3ade6dc19b6a510a3", "generation_seconds_batch": 18.26155725494027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.828247547149658, "target_z": 1.3515623270858519, "ce": 6.260893821716309, "genuineness": 4.344170093536377, "identity": 0.7120134830474854, "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_WARM_reg_high/e308ad6941509b563cfae50a3acae6a62fb3cc436d70b4835c64dbf7bfa01a61.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "e342f8cadd92f03be6135f99e7e72d78849ff15b58fc042f0674bb97a774f752", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "95e40807fbec5dd777f9656fdfb91e0e13f052b9618e99e78c923541cdc30dc9", "generation_seconds_batch": 24.12173402903136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.300607204437256, "genuineness": 3.9173994064331055, "identity": 0.34325850009918213, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e342f8cadd92f03be6135f99e7e72d78849ff15b58fc042f0674bb97a774f752.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "e37bfb34f99b5f159734dee320e7ee00c4d029ae4801f141fbf4f009fb760e3a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "db63ab84a3253085c957b04484d3037e71d094aab68fde3bfe3b9e99c2018df5", "generation_seconds_batch": 21.63054160901811, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7151408195495605, "target_z": 1.2265624588299646, "ce": 6.109982967376709, "genuineness": 4.093128204345703, "identity": 0.6315170526504517, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e37bfb34f99b5f159734dee320e7ee00c4d029ae4801f141fbf4f009fb760e3a.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "e41d329569592e7e138b96b8927e83e6742aa3d0e5ea454998eb1be055247f8a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "d980c5bcf7b93d975e746fd55a3aa465f351f3268113f08f283393d4d39abd9f", "generation_seconds_batch": 17.919459426077083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.983769655227661, "target_z": 1.5234375411700354, "ce": 5.982900142669678, "genuineness": 4.281409740447998, "identity": 0.6683525443077087, "hypothesis": "Ik 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_WARM_reg_high/e41d329569592e7e138b96b8927e83e6742aa3d0e5ea454998eb1be055247f8a.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "e454b8b054453b3afe513280f3ac764115b7c1f49a22df3603e5398b725f7869", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "86062cad578f061e132472c3ffb92cfbec823bafb9fc31e2cb891f64c0b3944f", "generation_seconds_batch": 16.321040828013793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4394428730010986, "target_z": 0.9218749505959577, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.733647346496582, "hypothesis": "ik 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_WARM_reg_high/e454b8b054453b3afe513280f3ac764115b7c1f49a22df3603e5398b725f7869.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "e4b04625257efbd9215ef6c77db544e038e826d56c19662550d0080ef83439ca", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "0f9df6f3ba9ad20fef00f5c5596158e663cfca669ad34d37b540b28c6d9a18e2", "generation_seconds_batch": 17.67146022093948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.332377910614014, "genuineness": 4.482243061065674, "identity": 0.20046906173229218, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e4b04625257efbd9215ef6c77db544e038e826d56c19662550d0080ef83439ca.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "e51aa8e6b794972ace00033e07cc952afe151a7eaa4ee376c10788178045b530", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b02587590cb56e439b487a8d00edb37e82fefa59333373dfb65332db29db397e", "generation_seconds_batch": 17.237253520986997, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2521097660064697, "target_z": 0.71484381998906, "ce": 6.23706579208374, "genuineness": 4.111956596374512, "identity": 0.28409504890441895, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e51aa8e6b794972ace00033e07cc952afe151a7eaa4ee376c10788178045b530.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "e56f444cf985810c8ed08535bf730601b13b5fd94f3c2c465eeac3cf7dd819a6", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6dbabcd9799edaea99ae4ef9a8680e05a9e740ca7dbc8bf9ea239e996627725f", "generation_seconds_batch": 23.352548515074886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.167568206787109, "target_z": 1.7265624588299646, "ce": 6.3641486167907715, "genuineness": 4.168440818786621, "identity": 0.5794931650161743, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e56f444cf985810c8ed08535bf730601b13b5fd94f3c2c465eeac3cf7dd819a6.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "e574584f6d25b4c0966aad3b8550a791a21dd51eb401e18b3c7c78ea54d9bb18", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "42bcc86f4ee1b80f34895ae90b4bea5ce514618bbfaba74168baa7332497e826", "generation_seconds_batch": 17.002740437048487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.006728172302246, "genuineness": 4.18099308013916, "identity": 0.606843113899231, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e574584f6d25b4c0966aad3b8550a791a21dd51eb401e18b3c7c78ea54d9bb18.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "e5887c35fbe86950fa4b82fb9a4bfecd3ef7a6d971d35c7ae58d64f1119d09ae", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4e3eca292bdf72b1d74dcb32bf99e053f7da6dd302e7e3c930dbe7d70df3611a", "generation_seconds_batch": 17.483919998048805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.332377910614014, "genuineness": 4.149612903594971, "identity": 0.7077270746231079, "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_WARM_reg_high/e5887c35fbe86950fa4b82fb9a4bfecd3ef7a6d971d35c7ae58d64f1119d09ae.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "e5918b8755c7f5dbff3853eac88b5d24c7115dcb200e716830b56f2dd0d10ffb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7ba91472adce2e28b4ac4dfb51296e5d7484d8e4014f6f2f47bebee7f9532f5c", "generation_seconds_batch": 24.758669962990098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.292664527893066, "genuineness": 4.093128204345703, "identity": 0.30776315927505493, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e5918b8755c7f5dbff3853eac88b5d24c7115dcb200e716830b56f2dd0d10ffb.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "e59fc439260f2bc97f5409826a08869415f8a6d7285bd7b8be8af15803b5fd1f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ec6c9262f3b8c34d5c24f1774e11097f1eaa5740df10725ee225f09a92794bc9", "generation_seconds_batch": 17.199293129961006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.038498878479004, "genuineness": 4.337893962860107, "identity": 0.6506175994873047, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e59fc439260f2bc97f5409826a08869415f8a6d7285bd7b8be8af15803b5fd1f.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "e5a0b37c637264bf298f07f82da8cc06532179ae22d69ebde13c010d6b88460e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e3b0d9bc8a31cce1c5bfbac34d2022562c46f1853f71dd1743c2fa20e8bc3f39", "generation_seconds_batch": 21.257038010982797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2311906814575195, "target_z": 1.7968748188518449, "ce": 6.348263263702393, "genuineness": 4.432034492492676, "identity": 0.7954143285751343, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e5a0b37c637264bf298f07f82da8cc06532179ae22d69ebde13c010d6b88460e.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "e5df88432a3ee7c7ec683b2d6f462699c12d77113313f8c30809be355439375a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "6910f8a71fb0843958a028be4580c04368b3e1b1962a927090ff7401b2f72b82", "generation_seconds_batch": 22.276874042931013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1425375938415527, "target_z": 0.5937501646801411, "ce": 4.318910598754883, "genuineness": 3.9048471450805664, "identity": 0.6170206069946289, "hypothesis": "Police Jones and State Pitsi, une preuvinte au VR, une preuve d'IQLS, on a par succès.", "wer": 1.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e5df88432a3ee7c7ec683b2d6f462699c12d77113313f8c30809be355439375a.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "e6ad8bb563c952a008f77f552945b58189d9949a66b564a48d9ce8c35bda99ed", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5c693d7e22f29f64c863c5341f627f1bb2e5276a0e168e286ac044ec9af95197", "generation_seconds_batch": 16.910009634913877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446512222290039, "target_z": 0.92968763997812, "ce": 6.300607204437256, "genuineness": 4.444586753845215, "identity": 0.5232408046722412, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e6ad8bb563c952a008f77f552945b58189d9949a66b564a48d9ce8c35bda99ed.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "e6b0641590e5b67faf7662578bb83675d13d9b28bdd6312eac3aca55ad124e83", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3a650ff6e2b7feb925c8152e255d42f99a687e40f5c90d0fe3e78fe8b7c8571e", "generation_seconds_batch": 23.441742299939506, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.202914237976074, "target_z": 1.7656251152760987, "ce": 5.967014789581299, "genuineness": 4.287685871124268, "identity": 0.6563988327980042, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e6b0641590e5b67faf7662578bb83675d13d9b28bdd6312eac3aca55ad124e83.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "e6e29e8bf80ede2aa05e07b9fd47decf7d51dc2d8d805621cd0f2e79d21b0401", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40c11fa2080623c12e14f1b20b610225dece078ba5e982409ca985e923361c76", "generation_seconds_batch": 16.057800026028417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467719554901123, "target_z": 0.9531249176599295, "ce": 6.022613525390625, "genuineness": 3.9299514293670654, "identity": 0.5006506443023682, "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_WARM_reg_high/e6e29e8bf80ede2aa05e07b9fd47decf7d51dc2d8d805621cd0f2e79d21b0401.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(very warm vocal delivery. 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": "3d92f7c7c769764cd2a0590b17b3e076d5b8bf9549dbf5524c79e58d05ea55d1", "generation_seconds_batch": 17.741529052029364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.38288950920105, "target_z": 0.8593750164680141, "ce": 6.427689552307129, "genuineness": 4.507347106933594, "identity": 0.2544962167739868, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We'd put it back after lunch.", "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_WARM_reg_high/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "e70f9240a20ba8b1f3bc5760f0cffb42b4c881f70762734282d299feb86aaf27", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c567860619706aed09eb741604c2f1351b215de5a4ba71265bbe81e7d56c11c9", "generation_seconds_batch": 22.377116161980666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.24500846862793, "genuineness": 4.319066047668457, "identity": 0.7133007049560547, "hypothesis": "La porte est une tenante manche, c'est Nénade. La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.4444444444444444, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e70f9240a20ba8b1f3bc5760f0cffb42b4c881f70762734282d299feb86aaf27.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "e71a0b006cff7c3c453919d9ae80718bd2093d94e59165fe9a11d2ea595c0295", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a812bebec928de5863aeb8b0d334fe2e09bc7425ba985a4385063500622bcc6b", "generation_seconds_batch": 16.84673763893079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.404097080230713, "target_z": 0.8828125576380493, "ce": 6.2767791748046875, "genuineness": 4.557555198669434, "identity": 0.5076088309288025, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e71a0b006cff7c3c453919d9ae80718bd2093d94e59165fe9a11d2ea595c0295.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "e75ab6221a6f7e1c6ca246840eeb3ab1c03d0f8bf86df3e2ab14609da2c3dc39", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "ab996c2414701dd2744652db6ecdb730e6c7c1725cb165720a7bcb86fa360ef8", "generation_seconds_batch": 25.214339010999538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.149696350097656, "genuineness": 4.080576419830322, "identity": 0.30611786246299744, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e75ab6221a6f7e1c6ca246840eeb3ab1c03d0f8bf86df3e2ab14609da2c3dc39.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "e7673d736580887b12e232604b7590d4aaf182ae0956db81711370de21bae8da", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7167f5b14de812b2cb4213ed7583e84e7dc3242754f263cdedbceb0aaac955ba", "generation_seconds_batch": 18.51516955893021, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.3641486167907715, "genuineness": 4.444586753845215, "identity": 0.35563892126083374, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e7673d736580887b12e232604b7590d4aaf182ae0956db81711370de21bae8da.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "e7c734801a472b6374366fc1855da6c14f51245adc3b7acc6a3fa4571ec2e9bb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "dfb2ec954701345f3d753920fd9501fbef329904270f1470a5d46483455ec22f", "generation_seconds_batch": 23.9261788809672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174637317657471, "target_z": 1.7343748847239013, "ce": 6.181467056274414, "genuineness": 4.7458367347717285, "identity": 0.6299914121627808, "hypothesis": "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_WARM_reg_high/e7c734801a472b6374366fc1855da6c14f51245adc3b7acc6a3fa4571ec2e9bb.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "e82e6a5b69a4faf06193e264a3d7e9722c54cc230310ae30a7764fb341c2fd5c", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3eb4c9c4e8120aca88f1c5db6939203472f726d6c100477edf609f6d8e613b0b", "generation_seconds_batch": 17.307742376928218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.075668811798096, "target_z": 1.624999868255887, "ce": 5.95112943649292, "genuineness": 4.143336772918701, "identity": 0.4828093349933624, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e82e6a5b69a4faf06193e264a3d7e9722c54cc230310ae30a7764fb341c2fd5c.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "434131c945fa103725ed0cce2a476df58dfcc1e4475d7648663ce51770ac670f", "generation_seconds_batch": 17.171640424989164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.785109996795654, "genuineness": 4.331617832183838, "identity": 0.523501992225647, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna, naast het raam, neem ze mee als je weet.", "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_WARM_reg_high/e87ac4538dcb329989741f5445374d7c3da1bfda0274b98b9e28ae4d01bba5ae.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "e936049215ef8efb6c3459a033bacc06f900f71acd33a72db3960a20fec677fb", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "223b61814cd17efe2ee3566b6426cafec4c8e78de0eca490258be24554be20eb", "generation_seconds_batch": 22.75128999003209, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.485681533813477, "target_z": 2.078125312892268, "ce": 5.816103935241699, "genuineness": 4.532451152801514, "identity": 0.7628132104873657, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e936049215ef8efb6c3459a033bacc06f900f71acd33a72db3960a20fec677fb.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(very warm vocal delivery. 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": "395a10ac60c7ce15082dfb90c9f653f877a8afe5ed43a7367781b9a93d2b1981", "generation_seconds_batch": 19.759389695012942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.393493175506592, "target_z": 0.8710936553089189, "ce": 6.41180419921875, "genuineness": 4.105680465698242, "identity": 0.5849598050117493, "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_WARM_reg_high/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "e9f5494ab3ffb3e92b614c5e52eb55b017689033a5533c996cea298175a4dd66", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "ca28267bbb6b026180d48f4502b11f15428916ee3ffcef98673f5f34ca91c459", "generation_seconds_batch": 21.257038010982797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.906008720397949, "target_z": 1.4375000658720565, "ce": 6.086154937744141, "genuineness": 4.030367851257324, "identity": 0.7276676893234253, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/e9f5494ab3ffb3e92b614c5e52eb55b017689033a5533c996cea298175a4dd66.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "ea10ecdcc12cd27f19fdc9443ed108d1571be3cf391f7c61a7dfb36d6cdfdd1a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0ab5494284051ba60f5f0e1b943b1937752441bacaf46ed6f28ac84f8fd903df", "generation_seconds_batch": 23.849725817912258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.689797878265381, "genuineness": 4.394378662109375, "identity": 0.7507374286651611, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in komen.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ea10ecdcc12cd27f19fdc9443ed108d1571be3cf391f7c61a7dfb36d6cdfdd1a.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "ea6d949edf2398e4053607609dfd3d57a44cb6b9a4f266fd6411e021c2dd64b1", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a8138ceaad1c561b0f636e3c3f847e2d504cf1f474338a9bcbf944a42dc62a8b", "generation_seconds_batch": 23.875661403057165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.197352409362793, "genuineness": 4.457139015197754, "identity": 0.6628572344779968, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ea6d949edf2398e4053607609dfd3d57a44cb6b9a4f266fd6411e021c2dd64b1.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "3108462dc2a3bca702844f821344750d0edeefd8268bc2a95d4df6cce6470e19", "generation_seconds_batch": 17.815294245956466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5172038078308105, "target_z": 1.0078124258939365, "ce": 6.205295085906982, "genuineness": 3.8797430992126465, "identity": 0.6977337002754211, "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_WARM_reg_high/ea8ea5ec570b09452d34fc5ebf070c8234454c32e8a391e827d3d9d9617c8135.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "eaeac4d63798879a8eab24879aced250df93fede9f3f7d17b258df41fcb08ad8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1000a62ec8052a52c5d39505351b1c6b32956c04fcf2f427c5d322c884363298", "generation_seconds_batch": 24.598235002020374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 6.332377910614014, "genuineness": 4.068024158477783, "identity": 0.23139187693595886, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/eaeac4d63798879a8eab24879aced250df93fede9f3f7d17b258df41fcb08ad8.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "eaec11151fd8eb5cf7eae8c2482d3b83efe859200b15a0a05be89f63066764f0", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "72bb02723774a0d8475a8a7df3d86d6f36550a9df54d01576663c451fe34afe1", "generation_seconds_batch": 27.359727213042788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.358436107635498, "target_z": 1.9375000658720565, "ce": 6.006728172302246, "genuineness": 4.406930446624756, "identity": 0.5715398788452148, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/eaec11151fd8eb5cf7eae8c2482d3b83efe859200b15a0a05be89f63066764f0.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "eb3dce3f8639f93d690ab91aa832472093938f546b0a2d01263ea616d5cd403b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "41339832cdac1420001b952e35d04f9eb911fae7e108bdf9e34e085b35492a1c", "generation_seconds_batch": 24.792616149992682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6515183448791504, "target_z": 1.1562500988080846, "ce": 6.38797664642334, "genuineness": 3.9299514293670654, "identity": 0.6208796501159668, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/eb3dce3f8639f93d690ab91aa832472093938f546b0a2d01263ea616d5cd403b.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(very warm vocal delivery. 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": "dee3118afa3f57178aa9823e4d835a5c34226e02bab56d8768d30f654894aa1c", "generation_seconds_batch": 17.401499514002353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.404097080230713, "target_z": 0.8828125576380493, "ce": 6.109982967376709, "genuineness": 4.111956596374512, "identity": 0.8158816695213318, "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_WARM_reg_high/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "ec46228dda11e42c40a0a3c174ae4efd4cb6b69e2fb2902df96e35905cfcc20e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "42419f74b5d12948b89adfa56de400e13d3f6845335cedb3dfa529750430763a", "generation_seconds_batch": 17.69070610392373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8706626892089844, "target_z": 1.3984374094259224, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.757686972618103, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ec46228dda11e42c40a0a3c174ae4efd4cb6b69e2fb2902df96e35905cfcc20e.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "ec49cea3b678175cda9ad9744edf019879a0842f3c4e577583ed2153547c5bb2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f0e6ac4ac5be684cd0483b710f07165cccae056b68d7527bd2cb17f2214f4bbd", "generation_seconds_batch": 23.797212996985763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6727256774902344, "target_z": 1.179687376489894, "ce": 5.95112943649292, "genuineness": 3.7291181087493896, "identity": 0.7993242740631104, "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_WARM_reg_high/ec49cea3b678175cda9ad9744edf019879a0842f3c4e577583ed2153547c5bb2.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "ecbe98d797fd46d1bae2ff3b90090867d4b0fcc8c2fcc3386c1a6b3ccd3b5638", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c5770d066f81ce40789c8d01ad2c9cca4d3cb906356f78e5f36c6a93d1e9ce45", "generation_seconds_batch": 21.242770769051276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.125868320465088, "genuineness": 4.13078498840332, "identity": 0.6132591366767883, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ecbe98d797fd46d1bae2ff3b90090867d4b0fcc8c2fcc3386c1a6b3ccd3b5638.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "ece3fba10e377da28060babd585376ddc510ef6f0386f152aba565080b6a4bfe", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "fc868b470b47c39f85455fc3a72074d9e387740b35e8eea16f0bdd4ce21fb8e5", "generation_seconds_batch": 17.039087084005587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.070269584655762, "genuineness": 4.306513786315918, "identity": 0.6894581317901611, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ece3fba10e377da28060babd585376ddc510ef6f0386f152aba565080b6a4bfe.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "ed08ee7d3522f8ff48a069dbb027b913edbaea3f2677c4c9b268cd08253ea135", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "70745fffca378109db4bfa5950ca0c0f2313ce8a29c810ce0102b6e4dd48340c", "generation_seconds_batch": 18.172895153984427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8565244674682617, "target_z": 1.3828125576380494, "ce": 6.09409761428833, "genuineness": 4.344170093536377, "identity": 0.5950226783752441, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ed08ee7d3522f8ff48a069dbb027b913edbaea3f2677c4c9b268cd08253ea135.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "ed1c1f47532cde3b4b1298eee181936fffa75a728bbf0f2f56cee7281359b248", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "312804f774b981cba0451bc46efb70146bc6d4c805773e4145d067bd34b0434b", "generation_seconds_batch": 19.710256452090107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4606504440307617, "target_z": 0.945312491765993, "ce": 5.919358730316162, "genuineness": 3.8671910762786865, "identity": 0.6439234018325806, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ed1c1f47532cde3b4b1298eee181936fffa75a728bbf0f2f56cee7281359b248.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "ed9aa3db29de15f052e49ddcb785def4b4c75150d0abe8333fa06eb27afae063", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "f7800548231a018d1f60970707385d1b2ca267a180640477e45d421693f86f9a", "generation_seconds_batch": 23.325892319902778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.195845127105713, "target_z": 1.7578126893821622, "ce": 6.205295085906982, "genuineness": 4.808597087860107, "identity": 0.7929075360298157, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ed9aa3db29de15f052e49ddcb785def4b4c75150d0abe8333fa06eb27afae063.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "ee2f318d4e2e11807b9219b19999caa4ac9d25cfcba233de7c60606a5bb7a30d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a7370d4d20772481b0a099a1165a263a4bca5708b3e3fe8115f4c4780143a4a3", "generation_seconds_batch": 23.409014391014352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.146360874176025, "target_z": 1.7031251811481551, "ce": 6.300607204437256, "genuineness": 4.507347106933594, "identity": 0.7189145088195801, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ee2f318d4e2e11807b9219b19999caa4ac9d25cfcba233de7c60606a5bb7a30d.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "685e498c3a9fa776cfde985cee2652eafbd72019d89128497dae7e975987cc58", "generation_seconds_batch": 21.184748630039394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7575559616088867, "target_z": 1.2734375411700354, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.47733259201049805, "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_WARM_reg_high/ee4c41998544204d4fcd21dc33abf5013557b2f29399dbc76acbef104eec3944.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "eea13615b9349cb9354341e6d7d899af6f727f95d929353d0f7b36d9324245f7", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "69df2631db73c90201aa17ff41b145da89b2d8ad4e5716005e54ebbc6fb8fdaf", "generation_seconds_batch": 24.521800426067784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.356205940246582, "genuineness": 4.268857479095459, "identity": 0.42529457807540894, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/eea13615b9349cb9354341e6d7d899af6f727f95d929353d0f7b36d9324245f7.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "eebaedc1c6a6d8002baabf928fd43db252bca4cc51a1a3041e35711ba03e5e9b", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "736a150c4010a15845891e21a1c56b17f2c0f9f0f6349eb67d85b20087a6a842", "generation_seconds_batch": 16.228829914005473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2768518924713135, "target_z": 0.7421875741060635, "ce": 6.332377910614014, "genuineness": 4.256305694580078, "identity": 0.5485751628875732, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/eebaedc1c6a6d8002baabf928fd43db252bca4cc51a1a3041e35711ba03e5e9b.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "eee18f8a1470aa8287c1e5bd1fa02d179f4317f34e9c2c0d287eef188859dba4", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2837f7fb7db939aee32aa6d4b9e15d89513a310f6aaa0f4c8021c9d723e9b27a", "generation_seconds_batch": 23.51501563191414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5808265209198, "target_z": 1.0781250494040424, "ce": 6.141753673553467, "genuineness": 4.218649387359619, "identity": 0.5316622257232666, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/eee18f8a1470aa8287c1e5bd1fa02d179f4317f34e9c2c0d287eef188859dba4.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "ef0608a00334f5fd15ad88de0a42a51dbcbca4de82bd10785fa210e35f34e4af", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7a3e8a23ac094584229ddd0b1baa55f8bf9672ee8637a72963549c27b66c9469", "generation_seconds_batch": 23.51501563191414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.708071708679199, "target_z": 1.2187500329360281, "ce": 6.372091293334961, "genuineness": 4.557555198669434, "identity": 0.6691718101501465, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ef0608a00334f5fd15ad88de0a42a51dbcbca4de82bd10785fa210e35f34e4af.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "ef24f730d7a82003a8632adeab65c04ce5dabd797323a47aa4273edde147d4c2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "e268a240f7f8517ad1c14f227e8488409e260cf42a16ed7f9351bd6d9dfdf1d1", "generation_seconds_batch": 18.52385436394252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.026184558868408, "target_z": 1.57031236002188, "ce": 6.356205940246582, "genuineness": 4.293961524963379, "identity": 0.32182127237319946, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ef24f730d7a82003a8632adeab65c04ce5dabd797323a47aa4273edde147d4c2.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "ef99f90e728d91ff24fbb526d131a895ea4e9daf514dc3c206dda785dccdfe94", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "9c9f075adad963cd763ae9282a5cf4104b0a3ea365d572da1b56d6e8b2865ebf", "generation_seconds_batch": 23.897229030029848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.459460258483887, "genuineness": 4.068024158477783, "identity": 0.6641337275505066, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ef99f90e728d91ff24fbb526d131a895ea4e9daf514dc3c206dda785dccdfe94.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "ef9d464cbd5eff31b223be767737d23b0e336eb34d64a07bd4f3934a73bd0244", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3faf71f9435e03c714e862cf2686b92ce321987efe48fb9362851f841d06cc5f", "generation_seconds_batch": 17.517801219015382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.4912309646606445, "genuineness": 4.545003414154053, "identity": 0.35280102491378784, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ef9d464cbd5eff31b223be767737d23b0e336eb34d64a07bd4f3934a73bd0244.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "f0a829fec90866e8154ae3a091058e5062ca7b0220584b63949cead55e4136ad", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1dc04d2418bf21000f9741624f8f93b2ab99d69cff5e9d784e932f632b3a9d9a", "generation_seconds_batch": 25.178820859990083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.068599700927734, "target_z": 1.6171874423619506, "ce": 5.621508598327637, "genuineness": 3.8169827461242676, "identity": 0.4871179759502411, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo.", "wer": 0.13333333333333333, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f0a829fec90866e8154ae3a091058e5062ca7b0220584b63949cead55e4136ad.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "f0b206b4ff2420a217de0339fc5dab3138687c13a8c24d205b777df7a52d514f", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "947262180d68ed3e5ab4aed071d8b0916af6f8410bddd8913414f0ae7192e47c", "generation_seconds_batch": 17.160005621961318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.24500846862793, "genuineness": 4.111956596374512, "identity": 0.7231190204620361, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f0b206b4ff2420a217de0339fc5dab3138687c13a8c24d205b777df7a52d514f.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "f12f1ae4f012f69603a0daac6c0c4a4688288db3e8a6b03521f984244a53a917", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "49f3605184d8bcd7b8a69b5fdf18bd3a24572d92f56d97dfd82c812990d9290c", "generation_seconds_batch": 27.55826400592923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.38003396987915, "genuineness": 4.168440818786621, "identity": 0.3035978674888611, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f12f1ae4f012f69603a0daac6c0c4a4688288db3e8a6b03521f984244a53a917.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "711861275eb063fe8c3193c0193b18a4ff70bfbfcd09f1b5bb16553b94a230f9", "generation_seconds_batch": 23.503982845926657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.38003396987915, "genuineness": 4.18726921081543, "identity": 0.7200809121131897, "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_WARM_reg_high/f142939c0568e0505e6845a274042c8ef86a9ee0f155375322e313ad22f6c1d3.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "f15d0d6122fda56ccdfad9d41dc4a600e573d263eccc7eca44fe22da932afa0a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "86d03504beb51bfc13d1e51adb8cabcb6160afc1331e482777882fdbf0c4513d", "generation_seconds_batch": 16.781978926970623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.157639026641846, "genuineness": 4.174716949462891, "identity": 0.6314736008644104, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f15d0d6122fda56ccdfad9d41dc4a600e573d263eccc7eca44fe22da932afa0a.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "f195180816d0b5400b20276c638fd75aedaf5b5d4cffb0ffebfd6872dd5ff312", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b0e78808c269131c5ee003aa1c4581353c5253302f92f21b4a4e8326a19de08d", "generation_seconds_batch": 22.168349427054636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.268836498260498, "genuineness": 4.432034492492676, "identity": 0.6681817770004272, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f195180816d0b5400b20276c638fd75aedaf5b5d4cffb0ffebfd6872dd5ff312.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "f1acf9325e11920825450e651a2d8dbd09ceeba651b422b83ac2968f915a4c08", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5b6fd11800b9232cf8bdf034e576dc842867ec8d4b22bed30d73d16ba40f4cde", "generation_seconds_batch": 23.58497692400124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.197352409362793, "genuineness": 4.262581825256348, "identity": 0.2778632640838623, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f1acf9325e11920825450e651a2d8dbd09ceeba651b422b83ac2968f915a4c08.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "f1b88883df7ec452a0abc576a602864830eb7a799aa164088944635ed43e7b22", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c5f06bf1813a07bb22273b335d023007a15e3f6332b41e57080f0ad158e53f28", "generation_seconds_batch": 16.668196739978157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.503065586090088, "target_z": 0.9921875741060635, "ce": 5.9431867599487305, "genuineness": 4.3127899169921875, "identity": 0.6592862010002136, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f1b88883df7ec452a0abc576a602864830eb7a799aa164088944635ed43e7b22.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "f1c693d5079e4010326f8f069c01e7c68d285808f8c1c8057d5fc555661453d5", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3e750c30761b039eae94315b74226bb2ae6c06605a693d502a440d8c9d30816b", "generation_seconds_batch": 17.059889992000535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.898939609527588, "target_z": 1.42968763997812, "ce": 5.935244083404541, "genuineness": 3.986435890197754, "identity": 0.5128337144851685, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f1c693d5079e4010326f8f069c01e7c68d285808f8c1c8057d5fc555661453d5.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "f2131c68bc294362c60dbf8a29b9e5d3234f7b5ea4ef0804d839429005e1fd6e", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7836664876704176d69f25ad11b06f23fd2b8a0451d0cbe176cfc01fe01d4ce1", "generation_seconds_batch": 17.21377755701542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.393493175506592, "target_z": 0.8710936553089189, "ce": 6.3641486167907715, "genuineness": 3.8295347690582275, "identity": 0.5796672105789185, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f2131c68bc294362c60dbf8a29b9e5d3234f7b5ea4ef0804d839429005e1fd6e.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(very warm vocal delivery. 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": "a0f3897f9f23e61f2140ee3b78e7c8bcf6020b892ee56bead0bae90906a5ed04", "generation_seconds_batch": 18.211912594968453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.149696350097656, "genuineness": 4.381826400756836, "identity": 0.7001606822013855, "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_WARM_reg_high/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "f223d0d85992180ac1ef78eb472c54ec3446dc32e156a8f50f45453229cc85d3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a86e0f19474503d02af6cb5a02ac2e5ce9d6be841698f5315495856be8f58b8d", "generation_seconds_batch": 24.598235002020374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.24500846862793, "genuineness": 3.8420867919921875, "identity": 0.33780866861343384, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f223d0d85992180ac1ef78eb472c54ec3446dc32e156a8f50f45453229cc85d3.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "f2600df20f804e6caf7ade6b6822bc1c9f1a19a14481a96f303e449aadb02e5a", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 126, "seconds_generated": 10.08, "mp3_sha256": "1e30112a01b89e0ce21fc04213e13f3655c3637f48c65668bd1298b873fa4ecf", "generation_seconds_batch": 21.678142830030993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.149606704711914, "target_z": 0.6015625905740776, "ce": 5.903473377227783, "genuineness": 4.218649387359619, "identity": 0.6963083148002625, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen", "wer": 0.1111111111111111, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f2600df20f804e6caf7ade6b6822bc1c9f1a19a14481a96f303e449aadb02e5a.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "f27164616f9cb4cb72dd7542034eca5cd249d42a2aa2f2f80439d5af60b73230", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fc6b83ea74034fb734ec76395c203b9357e016ba27a257b4f3a40ed0e24dc9f5", "generation_seconds_batch": 22.276874042931013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 4.739872455596924, "genuineness": 4.2374773025512695, "identity": 0.7054633498191833, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce quelconque.", "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_WARM_reg_high/f27164616f9cb4cb72dd7542034eca5cd249d42a2aa2f2f80439d5af60b73230.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "f3697a39559c1a534d148e27e4f98a4aa100ac69aea287199302fffe1a3e1b04", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "2a4a26c3d57665c425f65f761d8e8e3ca451930e1ba72d8f536815316c460b41", "generation_seconds_batch": 23.05154339398723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4076316356658936, "target_z": 0.8867187705850177, "ce": 6.141753673553467, "genuineness": 4.281409740447998, "identity": 0.6325105428695679, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f3697a39559c1a534d148e27e4f98a4aa100ac69aea287199302fffe1a3e1b04.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c448db5789d0e141a31e8174355526c0bc4615adcd43e821f9918cfb3f1de77b", "generation_seconds_batch": 17.336190436035395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8494553565979004, "target_z": 1.375000131744113, "ce": 6.125868320465088, "genuineness": 4.093128204345703, "identity": 0.647154688835144, "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_WARM_reg_high/f3855d03d05608079afd582126980f50f48b986917e923b3a195cefc800a0f23.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "f4001ff1b061fbdf41c8d1485682c855ecda212aba7ed484db5173735e4d59c9", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7ba9c477f382179872293ca622d54035bc1260c726f1764c49dbee60f498547f", "generation_seconds_batch": 28.555501414928585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.068599700927734, "target_z": 1.6171874423619506, "ce": 6.229123115539551, "genuineness": 4.720732688903809, "identity": 0.6252448558807373, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle pouvait.", "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_WARM_reg_high/f4001ff1b061fbdf41c8d1485682c855ecda212aba7ed484db5173735e4d59c9.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "f4dbbabd3387874e5b276b7edf4cd3c64caa84d48e344f70c3b483d565c17d30", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "f5ff245b1d227b8f2700662dc759a7de6acbb8f61e15a6791c27a370df5870cf", "generation_seconds_batch": 15.410423223976977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.488927125930786, "target_z": 0.9765624588299647, "ce": 6.181467056274414, "genuineness": 4.545003414154053, "identity": 0.7635293006896973, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f4dbbabd3387874e5b276b7edf4cd3c64caa84d48e344f70c3b483d565c17d30.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "f52203ac354c8b58dd11c0322c8e080e843de94d7a43c8c63b5aa942db87794e", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "c3aadf4c3355ff0774e8d8695af4c30e28520ba45b872d88a50ae4f8d201691d", "generation_seconds_batch": 22.671761453035288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 6.086154937744141, "genuineness": 3.8295347690582275, "identity": 0.5790876746177673, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f52203ac354c8b58dd11c0322c8e080e843de94d7a43c8c63b5aa942db87794e.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "f6509a1d214f7388c52c0a37ecccdfe334eb54ed5509bfa52e73d10ea4fdcb83", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3dd1a8c62285ec4638d19b8e983dadd04dba7949a1bca2e82c02e64a8290575c", "generation_seconds_batch": 17.08630894299131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5525498390197754, "target_z": 1.0468750823400705, "ce": 6.332377910614014, "genuineness": 4.193545341491699, "identity": 0.5400193929672241, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f6509a1d214f7388c52c0a37ecccdfe334eb54ed5509bfa52e73d10ea4fdcb83.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "f682c6750906cf91e8662ede305e34c132a1b883881b661014484e3bf93b9d87", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "3a590168b7d87f081bdadd4a52b40c7ecc2edf1fb6890b3f9bf7043863f99002", "generation_seconds_batch": 15.206619591917843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019115447998047, "target_z": 1.5624999341279435, "ce": 6.324435234069824, "genuineness": 4.394378662109375, "identity": 0.33799153566360474, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f682c6750906cf91e8662ede305e34c132a1b883881b661014484e3bf93b9d87.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "f68ab996505d975cf64b4b634311c957c07230e4aa5818912cc3579e47c33055", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "144f5564a1e0ca71ee8cece9daef8d30b09a99114667cacb7ba18cf7bd268364", "generation_seconds_batch": 22.377116161980666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4606504440307617, "target_z": 0.945312491765993, "ce": 6.086154937744141, "genuineness": 4.124508857727051, "identity": 0.7335501313209534, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f68ab996505d975cf64b4b634311c957c07230e4aa5818912cc3579e47c33055.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "f6a919e2b273a48b9b14caa12e8017a96c0d11d49b364fde89d08f7bf28e2130", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4b039ebfd4ac5d588bf69d1f2fb9d499f51f050943e39ebe1f2f96e2c6649f2e", "generation_seconds_batch": 22.197307275957428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.4912309646606445, "genuineness": 4.218649387359619, "identity": 0.6306657791137695, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f6a919e2b273a48b9b14caa12e8017a96c0d11d49b364fde89d08f7bf28e2130.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "f6c595839711d6ef81abcb9b2150d5fffc52c130689940aa549667c3c3b2a08c", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "865f0a68ecc3c9537fd79556742c95386a625f7c76cee31ff33878be0eecf967", "generation_seconds_batch": 17.00631051498931, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.983769655227661, "target_z": 1.5234375411700354, "ce": 6.308549880981445, "genuineness": 4.042920112609863, "identity": 0.23441025614738464, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f6c595839711d6ef81abcb9b2150d5fffc52c130689940aa549667c3c3b2a08c.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ca4a2c33c6edfb44e68f609a23a2cb4d37353396e5a87611ac7321c4025b90d", "generation_seconds_batch": 18.633507664082572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.23706579208374, "genuineness": 4.394378662109375, "identity": 0.6559370756149292, "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_WARM_reg_high/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_WARM_reg_high-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-1", "key": "f7bddf15aeb6cbbadfd847972aa005b39b36d00d3f90791b65a9d874b2146b76", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "65e9fa5d91f2f94bb7594e9a9333b6c4fcb6106d269635b38f603d64e54750b4", "generation_seconds_batch": 24.792616149992682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.38003396987915, "genuineness": 4.15588903427124, "identity": 0.7177364826202393, "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_WARM_reg_high/f7bddf15aeb6cbbadfd847972aa005b39b36d00d3f90791b65a9d874b2146b76.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "f820542dab99d3244214552bf358959288d9293a6d7727b9468f154747e3a82f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "7eb634081fe7a0535d70781a3bc12c0b684221e9914539eb86ec047480566318", "generation_seconds_batch": 18.56928878207691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.38797664642334, "genuineness": 4.425758361816406, "identity": 0.329758882522583, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f820542dab99d3244214552bf358959288d9293a6d7727b9468f154747e3a82f.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_WARM_reg_high", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(very warm vocal delivery. 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": "3ef0e8aa71f09ea446a5d65070c44ca46fa11256890287a267bdc9ccfc0f4933", "generation_seconds_batch": 18.985152882058173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.701002597808838, "target_z": 1.2109376070420916, "ce": 6.3641486167907715, "genuineness": 4.381826400756836, "identity": 0.5829388499259949, "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_WARM_reg_high/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "f8776cb956a38d6db40a4ac80d9823dc49eed8a0dbf0774c0ed5c5f682900c3f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "84f64b3a81f7740c401b6d773954c06ff2661c7956c751ef6e2b469de076adf0", "generation_seconds_batch": 21.281110243988223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4291276931762695, "target_z": 2.015624851787873, "ce": 5.895530700683594, "genuineness": 4.68307638168335, "identity": 0.6778390407562256, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f8776cb956a38d6db40a4ac80d9823dc49eed8a0dbf0774c0ed5c5f682900c3f.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "f87a7ddc091eaab300664bb2188ac52a25493f6cbdce77513292f9ce1f9d30bf", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a663b74316e434dbae5c9bec83d8c9c18e9380f7230f7601432a3bb5a2f41a94", "generation_seconds_batch": 19.710256452090107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.799971103668213, "target_z": 1.320312623510106, "ce": 6.173524379730225, "genuineness": 4.306513786315918, "identity": 0.5922409296035767, "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_WARM_reg_high/f87a7ddc091eaab300664bb2188ac52a25493f6cbdce77513292f9ce1f9d30bf.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "f94a259bec0acd01d3e7498cbbd88fc742c900d31338fe1fea5de8815415f058", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "28b25af6e851b8aec1064843af1e3887749c9c0909bc524ad6090432602e4b6c", "generation_seconds_batch": 16.62254815909546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.587895631790161, "target_z": 1.085937475297979, "ce": 6.165581703186035, "genuineness": 4.218649387359619, "identity": 0.7448703050613403, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f94a259bec0acd01d3e7498cbbd88fc742c900d31338fe1fea5de8815415f058.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "f97c6649533646be2fa5020a5ce4d3787deb7d6759e97b7265b04553303af421", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "282d5c00ba774446a17c0e328c96e835b56facddf19f9865aa31ecf257d515bc", "generation_seconds_batch": 15.892186530050822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.260893821716309, "genuineness": 4.494794845581055, "identity": 0.7279973030090332, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f97c6649533646be2fa5020a5ce4d3787deb7d6759e97b7265b04553303af421.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "f9d250fa4905d255b82293ff3fa20b5c5fc8abc5e46bcb1c1c1c36535a1e4814", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "39cf83cbc39a009b7bf86f79608fbd1bec8ebe64a5c92986bdd2b5b792b5d484", "generation_seconds_batch": 16.43219315900933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 5.982900142669678, "genuineness": 4.3127899169921875, "identity": 0.536553680896759, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f9d250fa4905d255b82293ff3fa20b5c5fc8abc5e46bcb1c1c1c36535a1e4814.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "f9d503ec1d7a16582a3abcc3bbd40afbb6037203603f9efca2fff7d2563c0474", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "39312c15c8946b242d8f2441e70cd8bf6e56f7cfcd25bdf44bbc6c0d7003c1de", "generation_seconds_batch": 21.63054160901811, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.920146942138672, "target_z": 1.4531249176599295, "ce": 6.24500846862793, "genuineness": 4.268857479095459, "identity": 0.5799602270126343, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/f9d503ec1d7a16582a3abcc3bbd40afbb6037203603f9efca2fff7d2563c0474.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "fa27a6a0f4c911ed018315b059929018ebe1e585e9af5a9d34aa62eddf3ed826", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8d8eb8aafc4370e0777e7112d3e539bb93c803d4e0f4fd16d3744362b84fed2f", "generation_seconds_batch": 23.000958855031058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467719554901123, "target_z": 0.9531249176599295, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.8467750549316406, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fa27a6a0f4c911ed018315b059929018ebe1e585e9af5a9d34aa62eddf3ed826.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "fa33835b0b39fb1717533884849aad005da0636a6a9cf674552faf59edbdff53", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "831b3c91ed868154a5b085f370f23ad55af7bca895cbaf0dd470510ee8761490", "generation_seconds_batch": 21.05181061697658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729279041290283, "target_z": 1.2421873106178378, "ce": 6.0146708488464355, "genuineness": 3.7040140628814697, "identity": 0.42886489629745483, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fa33835b0b39fb1717533884849aad005da0636a6a9cf674552faf59edbdff53.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "fb59893cc3594edaf8b71523e9712d8ba068862ac154ca5bdd90580e586628d2", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0ee2c412c54bfa3b269ccf8a39f6900007b738f1818766eeeb103b433109c57e", "generation_seconds_batch": 23.568191019934602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1322221755981445, "target_z": 1.6874998023838306, "ce": 6.0305562019348145, "genuineness": 4.18099308013916, "identity": 0.5881484746932983, "hypothesis": "De deur 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_WARM_reg_high/fb59893cc3594edaf8b71523e9712d8ba068862ac154ca5bdd90580e586628d2.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "fb75e9868f74d3be6434e8a6278be53b7e25020b7f1bb4d8e5ce8635ac933507", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "64dd23d28b7768a152ff589dc5ca50c6fb8bfcc277292c8c22066d8351d8df63", "generation_seconds_batch": 26.50251111097168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.24500846862793, "genuineness": 4.444586753845215, "identity": 0.7250632643699646, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fb75e9868f74d3be6434e8a6278be53b7e25020b7f1bb4d8e5ce8635ac933507.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "fb82af4b5f0f52725cb7454e034c93eb8ed558b68458bd737f7955d852090946", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b12b116c93190ccf62d13a5ab76ce34ce36f65627d1dc81bb515526a2fb7a581", "generation_seconds_batch": 17.360030864016153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.471542835235596, "target_z": 2.0624999341279437, "ce": 6.149696350097656, "genuineness": 4.319066047668457, "identity": 0.5227245688438416, "hypothesis": "Y encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de aquí.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fb82af4b5f0f52725cb7454e034c93eb8ed558b68458bd737f7955d852090946.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "fba73702470117a2d92c3a3346eb5cb7cf25ba10361f972d6317fff2701b0c02", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "8ab9393743017a386b046f8ca08b860fb273a624acd1309eb10b5175dd256b92", "generation_seconds_batch": 15.949976561008953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2627134323120117, "target_z": 0.7265624588299647, "ce": 6.2767791748046875, "genuineness": 3.898571252822876, "identity": 0.1836497038602829, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fba73702470117a2d92c3a3346eb5cb7cf25ba10361f972d6317fff2701b0c02.mp3" }, { "id": "vn_WARM_reg_high-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-1", "key": "fbed07f088046e802110bdab8aa31ca2384c270a2c8ed36bd29c47b711e063db", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a230163b7c30aca7160a6282b876622d9e358e5edfb75dad2b39f3a36d30b837", "generation_seconds_batch": 17.52722038794309, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743417739868164, "target_z": 1.2578126893821622, "ce": 6.133810997009277, "genuineness": 3.8797430992126465, "identity": 0.7526782751083374, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fbed07f088046e802110bdab8aa31ca2384c270a2c8ed36bd29c47b711e063db.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "fc376f15dcb97be49929cda55cc50105901688ece3e8e05113ee0c0aace2ceb3", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5d6f3c1d4870f17b1175d78e17f7072af200413be365495d7c4ef97346318800", "generation_seconds_batch": 18.204944938072003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 6.24500846862793, "genuineness": 4.306513786315918, "identity": 0.6755906939506531, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fc376f15dcb97be49929cda55cc50105901688ece3e8e05113ee0c0aace2ceb3.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "fc3a03923d03ee5cad2752ac2b16f487b96feb30a70d2a21c70037ccf9e2eca9", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "11f8a8d88a1b5415ce6619215d564509ccc3da62748e4859bd3d0ef8799f32ce", "generation_seconds_batch": 16.986312270979397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821178436279297, "target_z": 1.3437499011919154, "ce": 6.2767791748046875, "genuineness": 4.325342178344727, "identity": 0.6291636228561401, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fc3a03923d03ee5cad2752ac2b16f487b96feb30a70d2a21c70037ccf9e2eca9.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "fc3ea36d05eb00aa27da5a5a874acd669d98c80dc695ff907d96431b660a1087", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ab2b9d4e866f54d2ffe021f4a98d6da72592ea70ef89b61c4956d3170c74432f", "generation_seconds_batch": 23.534055243013427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 5.772419452667236, "genuineness": 3.8922953605651855, "identity": 0.5868275165557861, "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_WARM_reg_high/fc3ea36d05eb00aa27da5a5a874acd669d98c80dc695ff907d96431b660a1087.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "fc67f2a08c09cc3a63faa59085321b3e4797b2847da0cdad6ca61b4e169ef313", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "6ead75f0942b3eb95089e53d063d06abd69834e4676a2fcce75d4a0686c470fa", "generation_seconds_batch": 22.468655899981968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4853925704956055, "target_z": 0.9726562458829965, "ce": 5.847874641418457, "genuineness": 4.507347106933594, "identity": 0.819525957107544, "hypothesis": "J'ai trouvé la petite. J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle fait.", "wer": 0.2631578947368421, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fc67f2a08c09cc3a63faa59085321b3e4797b2847da0cdad6ca61b4e169ef313.mp3" }, { "id": "vn_WARM_reg_high-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-nl-0", "key": "fc8595a017885355e9c024aca20a10f5dee03243b42dc72769ce90bd63937280", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "a0fda2dd0df10e4e629f31da12e2e0e0d9fc679ff2f8cc6b6b69691a6384d645", "generation_seconds_batch": 23.568191019934602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8706626892089844, "target_z": 1.3984374094259224, "ce": 6.022613525390625, "genuineness": 4.2374773025512695, "identity": 0.5470480918884277, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fc8595a017885355e9c024aca20a10f5dee03243b42dc72769ce90bd63937280.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "fc9821ca5bf9a2cde47a2d26eaf20557350ced0fbd342bd5052e9312efa00bd0", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "5f4690315a7aae11b7ed8a8b59ffd7ec108c6f9e215adc7a7dee5de0e2b471ff", "generation_seconds_batch": 15.963248020969331, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.173524379730225, "genuineness": 4.406930446624756, "identity": 0.7077577114105225, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fc9821ca5bf9a2cde47a2d26eaf20557350ced0fbd342bd5052e9312efa00bd0.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_WARM_reg_high", "phase": "heldout", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "c04119f005a4b8196ee26a5b9a204b8c31dfef1557291d28bfa8d445e1d42db1", "generation_seconds_batch": 24.02243950893171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.146360874176025, "target_z": 1.7031251811481551, "ce": 6.23706579208374, "genuineness": 4.306513786315918, "identity": 0.601515531539917, "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": 0.120000000000001, "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_WARM_reg_high/fd8079cb99627aa33d602840a2dbbe70d52c8e35fc20368c32373df40ec0d235.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "fd9aa5ed2a438959908c0321e2ea974a842c8fec448b89372d68457c6c504f3d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "b0975ed49c7afb353af0722858a463fea905661122ffd18dd2c6b7c498b21236", "generation_seconds_batch": 25.01297877705656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.5874389410018921, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fd9aa5ed2a438959908c0321e2ea974a842c8fec448b89372d68457c6c504f3d.mp3" }, { "id": "vn_WARM_reg_high-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-de-0", "key": "fd9ae31409d6abbe2efa3507d5ac7a12a67ac204d9532b47716c0bfbc932447b", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "111d4c660ee3c42352cb2fe9fc58459f508480834e2c73140a34481ae7d84655", "generation_seconds_batch": 22.704661333933473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5313422679901123, "target_z": 1.0234375411700354, "ce": 6.24500846862793, "genuineness": 4.331617832183838, "identity": 0.5348972082138062, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fd9ae31409d6abbe2efa3507d5ac7a12a67ac204d9532b47716c0bfbc932447b.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "fdc6024e1c5f58e8c7571a02764328b9eb5376858306273a68f85434eb2a505d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "e772d5dcf31bfaff3362dd8f051c8778fedc1b76c81ec865914af805d1d5d208", "generation_seconds_batch": 26.514590451028198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 5.816103935241699, "genuineness": 4.068024158477783, "identity": 0.5289074778556824, "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_WARM_reg_high/fdc6024e1c5f58e8c7571a02764328b9eb5376858306273a68f85434eb2a505d.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "fdd8d39e46701660afc406b17bdfb6fa436479e2ae7487eb29f962e2c97c1b42", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "3ff86b57b5a00dc85f2d910e8f4d0a86ebfe9cc5dcd2b3b68c0da7567bb17916", "generation_seconds_batch": 14.758363711996935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.004977226257324, "target_z": 1.5468750823400705, "ce": 6.046441555023193, "genuineness": 4.406930446624756, "identity": 0.4610089361667633, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fdd8d39e46701660afc406b17bdfb6fa436479e2ae7487eb29f962e2c97c1b42.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "fe61a63a02d78e8c97b6b7ef2f3d58a3be5ac86f3c7aff75366f8f3c94d4d0e8", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "16252f672b6b1aeb282c6d4c732d4987fcfbb16c1c84a842def4639c803fff0b", "generation_seconds_batch": 21.91711525397841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.252951145172119, "genuineness": 4.281409740447998, "identity": 0.6227485537528992, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/fe61a63a02d78e8c97b6b7ef2f3d58a3be5ac86f3c7aff75366f8f3c94d4d0e8.mp3" }, { "id": "vn_WARM_reg_high-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-1", "key": "ff69ffa32de878dff5d73ce2b5ee193d73a80306492909dfa05f20fa7b0890ca", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "4ed0c262f7017b8559c1be1ab3cd0118d0b095ab17ccf6a0dc8920a48746b1d2", "generation_seconds_batch": 23.9261788809672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174637317657471, "target_z": 1.7343748847239013, "ce": 6.070269584655762, "genuineness": 4.425758361816406, "identity": 0.7311288118362427, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ff69ffa32de878dff5d73ce2b5ee193d73a80306492909dfa05f20fa7b0890ca.mp3" }, { "id": "vn_WARM_reg_high-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-0", "key": "ff88aa37baa5d6cd78faef7b21cd0ef86119b2de43ad9517bb922fe9dc8efea3", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "609fab99247433b5a74628e5d40b7a253eb4141219d8451c465bc007eafdb70e", "generation_seconds_batch": 19.67038721498102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9130778312683105, "target_z": 1.445312491765993, "ce": 6.260893821716309, "genuineness": 4.482243061065674, "identity": 0.6798343658447266, "hypothesis": "La porte 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_WARM_reg_high/ff88aa37baa5d6cd78faef7b21cd0ef86119b2de43ad9517bb922fe9dc8efea3.mp3" }, { "id": "vn_WARM_reg_high-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-1", "key": "ffac71abe6c7caddc20099e0edccccafe9f397affc6acbb536a65f029532b99f", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "1672997cdc9f357286697156f4003dd6c39057940f2acdc66b04391cfcfd09b5", "generation_seconds_batch": 23.85519957402721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8353168964385986, "target_z": 1.359375016468014, "ce": 6.41180419921875, "genuineness": 4.444586753845215, "identity": 0.39964595437049866, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ffac71abe6c7caddc20099e0edccccafe9f397affc6acbb536a65f029532b99f.mp3" }, { "id": "vn_WARM_reg_high-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-fr-1", "key": "ffb6ab0ce818ee1b04a3f7ca3d74e2087b097bb4bf2bc5b40981ce7a58c6e9af", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fd3bdd1fe28e2cf0248f9ba31ee72e111a73a9563a8e93f9ccb6b4e7ea0ee65d", "generation_seconds_batch": 15.801240448025055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.118083953857422, "target_z": 1.6718749505959576, "ce": 5.756534099578857, "genuineness": 4.444586753845215, "identity": 0.6807347536087036, "hypothesis": "I trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ffb6ab0ce818ee1b04a3f7ca3d74e2087b097bb4bf2bc5b40981ce7a58c6e9af.mp3" }, { "id": "vn_WARM_reg_high-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-es-0", "key": "ffbd9d2fcab1481aeff835f6111a96e045ef20b810757beef6a86cfbb407f610", "bucket": "vn_WARM_reg_high", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "61ee01e2adc673647843734e945e9b012670baa91fe84bc3f9f65a8514b196bf", "generation_seconds_batch": 16.61556583608035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.764625072479248, "target_z": 1.2812499670639719, "ce": 6.022613525390625, "genuineness": 3.9927117824554443, "identity": 0.48507797718048096, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ffbd9d2fcab1481aeff835f6111a96e045ef20b810757beef6a86cfbb407f610.mp3" }, { "id": "vn_WARM_reg_high-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_WARM_reg_high-en-0", "key": "ffd6d71b212fa59295ec70cbf4a256ed388cb1d5101629bb3f0bc8d980688c1d", "bucket": "vn_WARM_reg_high", "phase": "grid", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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(Very warm and personally engaged delivery, with kind resonance and a sympathetic interpersonal tone. 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": "231ee4f098da06f90a69ef3f83e9514f5caca17333df020b0c535737366d627d", "generation_seconds_batch": 17.16836826596409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.389958620071411, "target_z": 0.8671874423619507, "ce": 6.181467056274414, "genuineness": 3.7291181087493896, "identity": 0.3136550188064575, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_WARM_reg_high/ffd6d71b212fa59295ec70cbf4a256ed388cb1d5101629bb3f0bc8d980688c1d.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "e00694d94d73272fd1b638bd882ad3a73cbae5501f0d66b8dd322479f5dfbb96", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4535813331604004, "target_z": 0.9375000658720565, "ce": 6.332377910614014, "genuineness": 4.319066047668457, "identity": 0.46146273612976074, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/1059b086fb6d28a0ca4cc5e2.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "ab76dbd6813bd844487e7147b14e24087e5a909df5b02d69b3675af5a61fbf81", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4747886657714844, "target_z": 0.960937343553866, "ce": 6.268836498260498, "genuineness": 4.268857479095459, "identity": 0.3629557490348816, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/e59a2edd8bda239f428ac32f.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "809b1b8cff7f1916a2057b4ddaef648373e25e7192cd0e3d1a3240e771754e97", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4747886657714844, "target_z": 0.960937343553866, "ce": 6.260893821716309, "genuineness": 4.344170093536377, "identity": 0.5631886720657349, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/4a12bc00996285ae9ae955e3.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "b371bf731c064c884d85ff4a21ae91ec1e878fedb7141fe7ddcfb0dedbdeae3e", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4429774284362793, "target_z": 0.9257811635429259, "ce": 6.23706579208374, "genuineness": 4.2374773025512695, "identity": 0.35956037044525146, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/9d61682ea79584014cf8ff78.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "b6f47973d8ef2b4b9b551b31bc4817a5ba1c8c1116eca756ea31362442115828", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.284721851348877, "genuineness": 4.344170093536377, "identity": 0.4881823658943176, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/dd2e0b91dc76f3afe2335e91.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "c1a38f1acf5a71e20b081b19a20964f390ec8fe8e445a768196bcb40509409f1", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467719554901123, "target_z": 0.9531249176599295, "ce": 6.340320587158203, "genuineness": 4.400654315948486, "identity": 0.6455512046813965, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/73d31aa33f5e145315da9670.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "87f3f15014d4dcfb2738ad88c0f2e6aee1a340e1a4859c5e6d0b97ec982df36f", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.679795026779175, "target_z": 1.1875000658720565, "ce": 6.197352409362793, "genuineness": 4.582659721374512, "identity": 0.5310407876968384, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/972879ddd24d755fba803ace.mp3" }, { "key": "original-vn_WARM_reg_high-de-0-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "a1fe55c4202d96d287e3a5a07d42848fd2c77be7b0ce3427976a6fb64f3af688", "task_id": "vn_WARM_reg_high-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3192667961120605, "target_z": 0.7890623929579083, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.48118624091148376, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/0998a3b67d899c31f7b6d6eb.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "00f4aaf20a89ca9b589cb5d8e082fc8d237663f725a9ac38a76d5543c0b69f85", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.340320587158203, "genuineness": 4.13078498840332, "identity": 0.6140491962432861, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/cd2a595dd91c3bb8b02eaeab.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "5d44dcb3791fecf7adbb02c7fc0875efe0d8f2c98a6db2fc8c82e31855207293", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6727256774902344, "target_z": 1.179687376489894, "ce": 6.459460258483887, "genuineness": 4.18099308013916, "identity": 0.5987330675125122, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/3ecfbcf3948776142b431015.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "e397c8f4c92aacf9a8f82d5255ae829162bcb5feb3061256f9d1f88037838ce3", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.340320587158203, "genuineness": 4.256305694580078, "identity": 0.6139365434646606, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/997f8b2f94e69cdfd62937d0.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "03e147b645008744276ebbbf68e1a87cfc27e33042f2cdfcc64e13afcad2d92d", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7787632942199707, "target_z": 1.2968748188518449, "ce": 6.324435234069824, "genuineness": 4.15588903427124, "identity": 0.7276087999343872, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/b99e94d089b5173a414eb000.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "3bf18c03a78709e5fe26c6c30abed23c5cab8ba4f60e89af325a3a3e71e98e49", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4606504440307617, "target_z": 0.945312491765993, "ce": 6.38003396987915, "genuineness": 4.105680465698242, "identity": 0.4877181649208069, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/1533890555a137e8c5472096.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "420174fafd05d98d4c6b68fd3dbaff4b1822caebce6d9841e024bd21e6556e83", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.510134696960449, "target_z": 1.0, "ce": 6.38797664642334, "genuineness": 4.174716949462891, "identity": 0.5309768915176392, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/a803045ee872eea63ae6002c.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "43f1cc3ad25bc1de65ff5e1f22fab857d8b8bc88c0285cd2d4ba9fe4af77036e", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.332377910614014, "genuineness": 4.268857479095459, "identity": 0.6115773320198059, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/cf02f9d589f8962997c99308.mp3" }, { "key": "original-vn_WARM_reg_high-de-1-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "958d8f109fa296e023a7224f1c17d13361dbc706cde63f87409d5b96f1b2ac33", "task_id": "vn_WARM_reg_high-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.316492557525635, "genuineness": 4.2751336097717285, "identity": 0.5632142424583435, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/de/91974cf7fd7d4ca360198927.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "9f6223328c28ed1cd72dce4d53d7236b2aa20ee5ac480566edf57e9bd7925fed", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4429774284362793, "target_z": 0.9257811635429259, "ce": 6.141753673553467, "genuineness": 3.760498285293579, "identity": 0.29191410541534424, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/d7edbd1985604b7c3b504a8f.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "c420be8496a93747626461d262e399ad06fc70f8bf96b47d742df8bb2a76bed1", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4500465393066406, "target_z": 0.9335935894368624, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.0736958384513855, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/6d896e802040ad42aa18ebb1.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "f13c928add0c818d94d7e33ec3b4d0bf00ea42ce6f73756f0b0bb81c0921f66c", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.284721851348877, "genuineness": 3.9299514293670654, "identity": 0.1932157278060913, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/8ef4ab4b4618b4e14c48289f.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "4b79ac5422808390c1a18d78bcf1b9faeacd5778c6ec1d86aa2f991002aa529e", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2768518924713135, "target_z": 0.7421875741060635, "ce": 6.2767791748046875, "genuineness": 4.005263805389404, "identity": 0.18678413331508636, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/faf457e09a19c2bfe496cbf1.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "c9b1032fe912f2794db249687d3f72c2302cfb6c871f7e5aafda7fd83d4547a5", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446512222290039, "target_z": 0.92968763997812, "ce": 6.197352409362793, "genuineness": 3.8295347690582275, "identity": 0.4506334364414215, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/c8054e723939cd5280e1a1a9.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "12dfcc35aa616114d0aa7191cf61af85821b6d61953df49cf2f3bd8a12c1eeac", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3970279693603516, "target_z": 0.8750001317441128, "ce": 6.205295085906982, "genuineness": 3.9299514293670654, "identity": 0.29601380228996277, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/e19bdfc1e6adbbd37d048923.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "c36dd11678b385bca08907d7ee5c3be524cc7ec9fa82724e60bb53cecbacb78f", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2309021949768066, "target_z": 0.6914062788190247, "ce": 6.062326908111572, "genuineness": 3.873466968536377, "identity": 0.2437262237071991, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/21e559b28fe7e063051a8f79.mp3" }, { "key": "original-vn_WARM_reg_high-en-0-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "f6d618156beef2e8ec5b3dc260a10c46c88942016100d672c6fed2dd625fd59f", "task_id": "vn_WARM_reg_high-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7363483905792236, "target_z": 1.25, "ce": 6.2767791748046875, "genuineness": 4.381826400756836, "identity": 0.32266706228256226, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/c6a875355511e6a73337e507.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "d5fd5acb55d9393e80be03f390a0f9593defe9e313262e3f9d0ab781462b7dcc", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 6.340320587158203, "genuineness": 4.482243061065674, "identity": 0.49202531576156616, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/eea6d3e30fe5b076e13ffcb9.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "7345ad87b6c56613f8594e3691d1672b35453509ac6dbb8a291c9333756f91a8", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.644449234008789, "target_z": 1.1484376729141481, "ce": 6.3641486167907715, "genuineness": 4.607763767242432, "identity": 0.383414626121521, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/221502e3346db2db58372a41.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "476b2fb6f8052fcf2b4fd2e081a3a3d5d177158a12496716515bc8c3c525fc4c", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.4912309646606445, "genuineness": 4.469690799713135, "identity": 0.28412532806396484, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/76cefc2623996ccddf3ae9b3.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "d0325a82b33f6c5b35cfcb4fc16bb8fe26d596ab28e94ae685a15aa66f5bee21", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8777318000793457, "target_z": 1.406249835319859, "ce": 6.24500846862793, "genuineness": 4.494794845581055, "identity": 0.4262028634548187, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/1254d7b4f5785946aec2f9dd.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "89741f7d4c8c25f0532315ba8b664523c105ed1811381569328072b5ac548aa7", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.340320587158203, "genuineness": 4.20609712600708, "identity": 0.412373423576355, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/127954646ea07112bba02859.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "7fad95082a2016027fcf9bdab5d4feb92461a89729f42c6ca2fadab26dc6dd1b", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.701002597808838, "target_z": 1.2109376070420916, "ce": 6.324435234069824, "genuineness": 4.2249250411987305, "identity": 0.4562116861343384, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/b1f51a58699064ee483c8c10.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "f00760f835f400157a1bdbb8e7dbd180af58f24720c1930b6a15b6507b66d56f", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.2940455973148346, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/906f190b91ef7dbf06202dde.mp3" }, { "key": "original-vn_WARM_reg_high-en-1-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "2ff719a09cdd2bec2a6a94de3afbb5dd95db7b4ced254c3ddfd7d9f320e97070", "task_id": "vn_WARM_reg_high-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.623241424560547, "target_z": 1.124999868255887, "ce": 6.308549880981445, "genuineness": 4.306513786315918, "identity": 0.4773021340370178, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/en/f01272392bcfefa0caf238e3.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "6e3ebf6b4dcdcc2d40bd960df5a97f650094e64f8ff5ce12eb64d34f8a608be4", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7151408195495605, "target_z": 1.2265624588299646, "ce": 6.070269584655762, "genuineness": 4.111956596374512, "identity": 0.6031336784362793, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/7aa351a95151bd9ad3927424.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "57af9aa76537ff28759b3c5a594330e1f06b961f37716ba4abb5f9632a246075", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 5.95112943649292, "genuineness": 4.042920112609863, "identity": 0.5920372009277344, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/f18c98019ba3890dbd3d7496.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "122045c0016d959b3de518bb59378ce59552c709aa75f45d0073f2c372c205ea", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5737571716308594, "target_z": 1.07031236002188, "ce": 5.796247482299805, "genuineness": 3.948779582977295, "identity": 0.6611073017120361, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/203ec1f8d91f2994d5811f6e.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "1448d2848b94a6ea44e2a56e6451e2be6bb2cf51a2fc685393caade221b65552", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7504868507385254, "target_z": 1.2656251152760987, "ce": 6.078212261199951, "genuineness": 3.886019229888916, "identity": 0.46765202283859253, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/30c9bc2af4c3973532367761.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "fa843a6c3939048c2a658c8990576face131189dfa16b82e433d56d9c7e199aa", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4959964752197266, "target_z": 0.984375148212127, "ce": 6.173524379730225, "genuineness": 4.011539936065674, "identity": 0.6184074282646179, "hypothesis": "La puerta sigue abierta. He hiriado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/17dd63048642698a925b1090.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "87f18cfee7d4eb79d32837abc879f0a1b6c699262952717196e56b258a76fdb7", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6161723136901855, "target_z": 1.1171874423619506, "ce": 6.117925643920898, "genuineness": 3.7793264389038086, "identity": 0.561408281326294, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/8714656699aadb5703ca579e.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "eb8d1bcedc8f4a4a1b5137fce55101a9019c623d5495c4bcfe252dc865c25061", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 5.95112943649292, "genuineness": 3.7416701316833496, "identity": 0.48439979553222656, "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/audio/vn_WARM_reg_high/es/0a54d7047ce6ad3a41a18faf.mp3" }, { "key": "original-vn_WARM_reg_high-es-0-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "a7ec39f87197df1b000b21af8ef793f86f25428a90b1ed7a5888c27f01d2f976", "task_id": "vn_WARM_reg_high-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.722209930419922, "target_z": 1.2343748847239013, "ce": 6.0146708488464355, "genuineness": 3.9425034523010254, "identity": 0.42777198553085327, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/010f98f83582dbd97f583ce7.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "556788501487f1b9c588de100754a3132e7b1fbdb375b48f36fa4be2f294c161", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 6.086154937744141, "genuineness": 4.319066047668457, "identity": 0.47550684213638306, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/f8fd0c667688f5c8a1317786.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "934dc97eb21fcf71901ce4d4e97b9c617aa4100c0ad48b26bb862909dc0fd3fe", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.181467056274414, "genuineness": 4.532451152801514, "identity": 0.48748767375946045, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/d987908ae09699f33624478c.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "51dedc041a235764e33a413a43b1d8eda5341f71295aae43b0f335fcc641e63d", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9908387660980225, "target_z": 1.5312499670639719, "ce": 6.24500846862793, "genuineness": 4.582659721374512, "identity": 0.4152287244796753, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/8067a0332c77f6bebd7c1831.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "c98a30ac3880dc1302da41b7146eeaa21fce7d265ce33105a72e4c873744dbe5", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.807040214538574, "target_z": 1.3281250494040424, "ce": 6.086154937744141, "genuineness": 4.595211505889893, "identity": 0.7229107618331909, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/5beeed75d731f872c7e9505a.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "ad6a70876d3d1d03994e329734ed059ce73cd312dfd159f94a88b8f595461dac", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7929017543792725, "target_z": 1.3124999341279435, "ce": 6.078212261199951, "genuineness": 4.7332844734191895, "identity": 0.778176486492157, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/ebd29e9b3e674aacb41c28ef.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "219aa8a9b051dd39f507362ec113f53eed4f6550c14a12da0b1288922e503316", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 5.990842819213867, "genuineness": 4.582659721374512, "identity": 0.5498063564300537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/52cdfc003e14fed599f2d57c.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "e186e3b3eb5e33d7095f846e0c5701ce515e7ddaa9e523576b2549c8f586d167", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061530590057373, "target_z": 1.609375016468014, "ce": 6.062326908111572, "genuineness": 4.256305694580078, "identity": 0.8379853963851929, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/348e742209e06bb602d36484.mp3" }, { "key": "original-vn_WARM_reg_high-es-1-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "fd5fa6eea1c41e9563cf9463cda64ba90de7dc52b72ac16cdc84a5155cc93a60", "task_id": "vn_WARM_reg_high-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.252951145172119, "genuineness": 4.607763767242432, "identity": 0.6186424493789673, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/es/2c617920025b28bc91442d3b.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "9e44531e48966a8273a8cf039e32454e0cab1e64ba5202a509f9246d2a36acbe", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2311906814575195, "target_z": 1.7968748188518449, "ce": 6.157639026641846, "genuineness": 4.356721878051758, "identity": 0.7294650673866272, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/42c1d69d31a2a2ed6f3f018f.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "8d286535edc45ee96a8575f622581bb84708778091f1829dcce76f4224a4ca40", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863593578338623, "target_z": 1.390624983531986, "ce": 6.284721851348877, "genuineness": 4.607763767242432, "identity": 0.7094969749450684, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/53090a7b20991bf24aa73ae1.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "477e2d64945d179dfb1bc80280411ce310c97c6ad23d716158dc5f04d32ee3d4", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.221180438995361, "genuineness": 4.149612903594971, "identity": 0.8793678283691406, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/0bfc8629494f225a8fea0b50.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "8265215fe2d48cf67175308b050226a752c7f9e983a90a77f59dd1af845e7bbc", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9342854022979736, "target_z": 1.4687500329360281, "ce": 6.149696350097656, "genuineness": 4.13078498840332, "identity": 0.7979799509048462, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/7ca9f74be549b9cc4683077d.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "02ea8c074cf0fce544078f1d8d3168682a20a97856a4cd30304775c9fa9ad137", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.927216053009033, "target_z": 1.460937343553866, "ce": 6.197352409362793, "genuineness": 4.168440818786621, "identity": 0.6787395477294922, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/fa29f59d2a6ad9cf4e5ab791.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "becca53105332c6d8be332f2cc62ebfac2c3ed436c4667f440ff69b47503e3a3", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.6826356649398804, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/deb5e33d3d645b421d8e4e9f.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "1ed91b1b37046b2c1e933fd5a4036fd527abcbe55eee6b3a87a8996f314ecd38", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.229123115539551, "genuineness": 4.595211505889893, "identity": 0.747008204460144, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/70acc7a80e44638eb9b7f487.mp3" }, { "key": "original-vn_WARM_reg_high-fr-0-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "3420bb5d2abd498dd9401b91a5a21ad3f0543ec16399a552cd04314e31df92d6", "task_id": "vn_WARM_reg_high-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.609103202819824, "target_z": 1.109375016468014, "ce": 6.0146708488464355, "genuineness": 4.030367851257324, "identity": 0.7477089762687683, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/deafa592cefd8d3241f3c80f.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "4d533760bbae4f6ba53bd89ebfd75f8424f671080b4ed8ed3046a12b27cd2144", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. 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": 4.245328903198242, "target_z": 1.8124996706397178, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.7626249194145203, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/95321fb414ef8c959496a95e.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "23af95a233b87ee757ff53611e69bda52187e3516f88dab2c4583607d7957445", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9484238624572754, "target_z": 1.484375148212127, "ce": 6.24500846862793, "genuineness": 4.3127899169921875, "identity": 0.7527454495429993, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/cac615d3609fc95f422cd663.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "87a3480e84d9c9357b0f4ca96dd495885c004064194d9d9bb19cdd27caa79540", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. 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": 4.153429985046387, "target_z": 1.7109376070420916, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.7143789529800415, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/c378b46b98f37c0e0c3cc073.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "6c90b9d001438cc41831c0a7d7a101842a81f785874a152ac5a0e6f5fb62e2b1", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. 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": 4.026184558868408, "target_z": 1.57031236002188, "ce": 6.109982967376709, "genuineness": 4.20609712600708, "identity": 0.7329380512237549, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/6dbf1dee4225c06cbd141dbf.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "17089f3dfb11ad87f615a6832d31592841af0c75a869a2e7bdca90ea5ad35df0", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918702602386475, "target_z": 1.4218749505959576, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.7322866320610046, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/d3f078c15d4fc90bde00725d.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "6a5c2a81d4bff32e937922e047b59656de96600bd3913ab8cb6beaf3d6b7d429", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. 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": 4.04739236831665, "target_z": 1.593750164680141, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.7439266443252563, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/32c81f0471d78f56eaab6ce7.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "b070e143ad9fefdcaab28cbe42d2683a0c520a672167d7e7ede2604e97908ce4", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.962562084197998, "target_z": 1.5, "ce": 6.23706579208374, "genuineness": 4.808597087860107, "identity": 0.7642739415168762, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/8bf2abddf2acb0600cddb969.mp3" }, { "key": "original-vn_WARM_reg_high-fr-1-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "19e944273898e288d08a2cff114dda04cc774c03a1dc218685edaff692b4f935", "task_id": "vn_WARM_reg_high-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(very warm vocal delivery. 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": 4.082737922668457, "target_z": 1.6328122941498235, "ce": 6.125868320465088, "genuineness": 4.250029563903809, "identity": 0.6676735877990723, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/fr/0f41c572375958ca6f15b514.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "f424a1b773f5d541213b8687a9c9f0b5247f1733e624405ee01f18a11958bd25", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.492461681365967, "target_z": 0.980468671776933, "ce": 6.062326908111572, "genuineness": 4.2249250411987305, "identity": 0.6843066215515137, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/bcf32b158888db330394a8c8.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "6892b36d427cabf4b436354f29d38184d29a341ab1d20b4773597fe13d16d352", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446512222290039, "target_z": 0.92968763997812, "ce": 6.205295085906982, "genuineness": 4.300237655639648, "identity": 0.5550695657730103, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/febaabd44bc9beccb16639e0.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "51a6455969fb895c1e83fb6576087f11073d6e59974f87bd64cea4333450f88d", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6585874557495117, "target_z": 1.164062524702021, "ce": 5.959072113037109, "genuineness": 4.369274139404297, "identity": 0.7266964912414551, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/3924ff13fc003922adfddb73.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "ecb5bc1a0b4daaebeaa589c23ff6d627b1d5ab38ba90176bcb012ffa0b2bb623", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5949649810791016, "target_z": 1.093750164680141, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.7127708196640015, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/39385f9879a15af1def39132.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "da90f3b98e547310e1ef1a7dab7d716932ed0f4e7444f56816d1fa966e5620d2", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3440089225769043, "target_z": 0.8164061470749118, "ce": 6.078212261199951, "genuineness": 4.293961524963379, "identity": 0.7212189435958862, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/bce9621729be893a89423971.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "f1a7d163ac4684b24454569c92780596af9cc918446947c941e2476a3c17a342", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.372285842895508, "target_z": 0.8476563776271093, "ce": 6.062326908111572, "genuineness": 4.193545341491699, "identity": 0.5680782198905945, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/2bc9be4f5e359a081f5c55f4.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "f58af3958216f1ba048b790f420e67868fc62636428d01541f6b0b421eb1e9cd", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.379354953765869, "target_z": 0.8554688035210458, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.6222813129425049, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/f4c9bd392492c1232aed664c.mp3" }, { "key": "original-vn_WARM_reg_high-nl-0-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "fa611e2d41dd80813a06df20d948e7f8b37b620725447317f727538257eeb841", "task_id": "vn_WARM_reg_high-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4747886657714844, "target_z": 0.960937343553866, "ce": 6.109982967376709, "genuineness": 4.262581825256348, "identity": 0.5950008034706116, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/8563512f1f210b172f31629b.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1701-scale0.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "3db1de584987ff2d082caf3fb4a47c152b0fb96775b77a61dc05a7070f22a31b", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6939334869384766, "target_z": 1.2031251811481551, "ce": 6.125868320465088, "genuineness": 4.168440818786621, "identity": 0.6569207906723022, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/f3d335552cbf7c88721533b0.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1701-scale0.5", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "9f4819d226af25fbe25fd4457e07794058056a5f58c15fb72743af67330d42b5", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.499530792236328, "target_z": 0.9882810976708695, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.479899525642395, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/fb42e0b5aebfd6e52336a063.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1701-scale0.75", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "48223e9a6c281a0fb966254ea4c4715cb55beeb49b797528595f0a4237f558fc", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3546128273010254, "target_z": 0.8281250494040423, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.5303155779838562, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/16a9f0cd6bb1dd2b0532634f.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1701-scale1.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "1e05e0276187500a6a2862a712caba5c1e143efa84bd304e177489b0cd6730e0", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.785832643508911, "target_z": 1.304687508234007, "ce": 6.1020402908325195, "genuineness": 4.2751336097717285, "identity": 0.45361918210983276, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/5d4ae3f0fccddd72ab2f5510.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1702-scale0.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "d058cd6c60b54661914c701f4e801260ecf3b029ba1e0b952cdf8f3231676fe7", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566688060760498, "target_z": 1.0624999341279435, "ce": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.735770583152771, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/d2fb5fe4b3f670f84386d1c6.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1702-scale0.5", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "a58dc14e933d77389c548957471936baeacc97b70db6b4d5a5f3818a30011d9e", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7151408195495605, "target_z": 1.2265624588299646, "ce": 6.054384231567383, "genuineness": 4.256305694580078, "identity": 0.42223817110061646, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/3840f1228de248882396d627.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1702-scale0.75", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "3c6f25179438d41420b722b39538d4c486e10c87149124728d69aaa00d5cbc03", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.799971103668213, "target_z": 1.320312623510106, "ce": 6.165581703186035, "genuineness": 4.344170093536377, "identity": 0.81816565990448, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/5ca9b08d6ce0d6e135a86a6c.mp3" }, { "key": "original-vn_WARM_reg_high-nl-1-seed1702-scale1.0", "bucket": "vn_WARM_reg_high", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "bb1324e41fcaaabe1a7922ed59eebcd6d99725cceae9360ac398a73c54856d22", "mp3_sha256": "c171950b333cc3cf922b82ea5bc3185635079dc0a51a0f32f7d094418a9c9c29", "task_id": "vn_WARM_reg_high-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(very warm vocal delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.524272918701172, "target_z": 1.015624851787873, "ce": 6.308549880981445, "genuineness": 4.319066047668457, "identity": 0.5184311866760254, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_WARM_reg_high/nl/fc46a6e3870fc8fc30f9934e.mp3" } ], "source_standard_audio_unchanged": true }